INTEGRATING GENOMIC SELECTION AND GENOME EDITING STRATEGIES TO 

ACCELERATE POTATO BREEDING 

By 

Felix Eugenio Enciso Rodriguez 

A DISSERTATION 

Submitted to 

Michigan State University 

in partial fulfillment of the requirements 

for the degree of 

Philosophy 

2019 

Plant Breeding, Genetics and Biotechnology – Crop and Soil Sciences – Doctor of 

 

 

 

 

 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

 
 

 

 
INTEGRATING GENOMIC SELECTION AND GENOME EDITING STRATEGIES TO 

ACCELERATE POTATO BREEDING 

ABSTRACT 

By 

 

 

 
As a staple food, the potato (Solanum tuberosum L.) plays an important role in human nutrition 

Felix Eugenio Enciso Rodriguez 

and it is currently the third most important food crop after rice and wheat. However, the potato 

crop faces high production losses caused mainly by biotic factors. With the advent of cutting-

edge technologies suitable for potatoes, there is an increasing possibility to accelerate genetic 

progress and variety generation. To contribute to the implementation of genomic strategies to 

accelerate potato breeding, three different approaches were used. First, whole genome 

regressions were conducted using additive and dominant allele dosage models for late blight and 

common scab resistance in tetraploid potatoes. Multiple Single Nucleotide Polymorphisms 

(SNPs), contribute to late blight resistance, uncovering the introgression history for this trait 

whereas an unreported locus with a sizable contribution to common scab resistance was detected. 

Prediction accuracy assessments demonstrated that 90% of the genetic variance could be 

captured with an additive model, demonstrating the applicability of genomic prediction for 

tetraploid potato breeding. Second, a genome editing approach was implemented to breakdown 

the S-RNase-based self incompatibly in diploid potatoes. New S-RNase allelic variants, with 

flower-restricted expression, were identified in two self-incompatible (SI) diploid potatoes and 

mapped to chromosome I in a low recombination region. A dual single-guide RNA strategy was 

used to generate S-RNase knock-out lines producing premature stop codons on each targeted S-

RNase allele. Self-compatibility was achieved in T0 knock-outs and stable transmitted to T1 lines. 

Additionally, Cas-9 free plants were also obtained. Plasticity in the self-compatible response was 

 

 

also observed in wild-type lines, presumably associated with non-stylar and environmental 

factors. Third, validation of the IPI-O4-mediated suppression of the RB-based late blight 

resistance was conducted using in vivo and in vitro approaches. The hypersensitive response 

(HR) was confirmed when IPI-O1 was co-infiltrated with the RB gene from Solanum 

bulbocastanum using a heterologous system. However, HR was observed when IPI-O1 and IPI-

O4 were infiltrated in transgenic potato lines carrying a synthetic RB gene containing a Coiled-

Coil (CC) domain from S. pinnatisectum. Further work should be conducted to confirm this un-

reported interaction. Similarly, we could not validate CC-dimerization using yeast-two hybrid 

assays and therefore more extensive experiments should be conducted to confirm this result. 

Ultimately, these genomic approaches open a new window to accelerate the generation of new 

potato varieties. Genomic selection strategies along with targeted mutagenesis will expand the 

boundaries of both approaches, reducing the potato breeding cycle considerably while 

maintaining genetic diversity, and providing access to genomic regions with low or null 

recombination in potatoes. 

 

 

 

 

 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Copyright by 
FELIX EUGENIO ENCISO RODRIGUEZ 
2019

 

 

 

 
 

 

 

 

 
 

 

 
 
 
 

 

 

 

 

 

Dedicated to my Grandfather 

 

 
v 

ACKNOWLEDGMENTS 

 
 
 

I want to offer my special thanks to Dr. David Douches, for his mentoring and valuable 

constructive suggestions during the planning and development of this research. 

I would also like to thank my graduate committee, Drs. David Douches, Gustavo de los Campos, 

Robin Buell and Noah Rosenzweig for accepting being part of my Ph.D. committee and their 

advice and assistance. Especially, I would like to thank Dr. Gustavo de los Campos for his 

valuable contribution and support for the development of the genomic selection project. 

Similarly, I would like to thank Dr. Robin Buell for her guidance and valuable contribution to the 

self-incompatibility project. 

I want to extend my thanks to Drs. Swathi Nadakuduti, Daniel Zarka, Norma Manrique-

Carpintero, and Mr. Joseph Coombs, for their valuable help and contribution to each aspect of 

my dissertation projects. 

I wish to thank to various potato people from the MSU potato breeding program; Chen Zhang, 

Kate Shaw and Matt Zuehlke for their valuable support in the greenhouse and laboratory; 

Monica Hufnagel and Bernard Schroeter, for their assistance in tissue culture. I wish to thank 

Chen Zhang, Andrea Garcia, Swathi Nadakuduti, Norma Manrique-Carpintero, Maher 

Alsahlany, Kate Shaw, Natalie Kaiser, Donna Kells, Thylani Jayakody, Kelly Zarka, Juan 

Sandoval, Julio Valenzuela, Sandra Hernandez, Gina Garzón, Jaime Osorio and Yaneth Camargo 

for their friendship and support. 

I would like to thanks to Colciencias Fulbright – Fulbright for sponsoring me, and the Colombian 

Corporation for Agricultural Research (AGROSAVIA) for providing me permission to pursue 

 

 
vi 

my Ph.D., especially to Dr. Luz Stella Barrero for her support and encouraging me to continue 

my graduate program. 

Finally, I would like to thank my family for all their valuable support to make this dream 

possible. 

 

 

 
vii 

TABLE OF CONTENTS 

 
 

LIST OF TABLES ..................................................................................................................... xi 

 

LIST OF FIGURES ................................................................................................................. xiii 

 
Chapter 1 ......................................................................................................................................... 1 
ADVANCES IN GENOMIC STRATEGIES AND APPLICATIONS IN POTATO BREEDING1 
I. Potato origin and domestication ............................................................................................... 1 
II. Potato genetics and genomics-based resources ...................................................................... 2 
III. Vision for diploid potato breeding – Advantages and Challenges ........................................ 5 
i. Self-incompatibility in diploid potatoes ............................................................................... 7 
IV. Current production constraints in cultivated tetraploid potatoes ........................................ 10 
i. Common scab (Streptomyces scabies Thaxter) .................................................................. 11 
ii. Late blight [Phytophthora infestans (Mont.) de Bary] ...................................................... 12 
iii. Late blight-potato arms race ............................................................................................. 12 
iv. Genetic architecture of late blight resistance in potatoes ................................................. 13 
v. RB-mediated resistance in potato ...................................................................................... 14 
V. Genomic approaches apply to potato breeding ..................................................................... 16 
i. Whole genome association (GWA) analysis in potato ....................................................... 16 
ii. Genomic selection (GS) in potato ..................................................................................... 18 
iii. Genome editing: An efficient method to target agronomic-related traits ........................ 20 
a. Targeted mutagenesis technologies ............................................................................... 21 
b. Genome editing in potato .............................................................................................. 23 
VI. Project objectives ................................................................................................................ 24 
LITERATURE CITED .............................................................................................................. 26 

 
Chapter 2 ....................................................................................................................................... 42 
GENOMIC SELECTION FOR LATE BLIGHT AND COMMON SCAB RESISTANCE IN 
TETRAPLOID POTATO (Solanum tuberosum) .......................................................................... 42 
I. Abstract .................................................................................................................................. 43 
II. Introduction ........................................................................................................................... 44 
III. Materials and Methods ........................................................................................................ 46 
i.  Data ................................................................................................................................ 46 
ii.  Statistical analyses ......................................................................................................... 49 
a. Prior distributions for effects ......................................................................................... 51 
b. Variance components. ................................................................................................... 52 
c. Identification of SNPs with a sizable contribution to additive variance. ....................... 52 
d. Prediction accuracy evaluation. ..................................................................................... 53 
e. Software ......................................................................................................................... 54 
IV. Results ................................................................................................................................. 54 
i.  Variance Components Estimates ................................................................................... 56 
ii.  Cross-validation analysis ............................................................................................... 60 
V. Discussion ............................................................................................................................. 62 
APPENDIX ............................................................................................................................... 70 

 

 
viii 

LITERATURE CITED .............................................................................................................. 78 

 
Chapter 3 ....................................................................................................................................... 84 
OVERCOMING SELF-INCOMPATIBILITY IN DIPLOID POTATO USING CRISPR-CAS9
 ....................................................................................................................................................... 84 
I. Abstract .................................................................................................................................. 85 
II. Introduction ........................................................................................................................... 85 
III. Material and Methods .......................................................................................................... 88 
i. 
Plant material ................................................................................................................. 88 
ii.  Allelic identification, annotation and phylogenetic analysis of S-RNase ...................... 89 
iii.  S-RNase linkage mapping .............................................................................................. 90 
iv.  sgRNA identification, assembly and validation ............................................................ 91 
v.  Agrobacterium-mediated transformation ...................................................................... 91 
vi.  Molecular characterization of KO lines ......................................................................... 92 
vii.  Assessment of self-compatibility .................................................................................. 92 
S-RNase expression analysis ..................................................................................... 93 
viii. 
IV. Results ................................................................................................................................. 93 
i. 
Identification of the S-RNase gene in potato ................................................................. 93 
ii. 
Identification of allelic variants of S-RNase in diploid potato lines .............................. 96 
iii.  S-RNase is located within a pericentromeric region of chromosome I in potato .......... 97 
iv.  Targeted mutagenesis of S-RNase in diploid potatoes using CRISPR/Cas9 results in 
self compatibility ................................................................................................................... 99 
V. Discussion ........................................................................................................................... 105 
APPENDIX ............................................................................................................................. 110 
LITERATURE CITED ............................................................................................................ 118 

 
Chapter 4 ..................................................................................................................................... 125 
VALIDATION OF THE RB-MEDIATED RESISTANCE SUPPRESSION BY THE IPI-O4 
EFFECTOR FROM P. infestans IN POTATO ........................................................................... 125 
I. Abstract ................................................................................................................................ 126 
II. Introduction ......................................................................................................................... 127 
III. Material and Methods ........................................................................................................ 129 
i. 
Plant material ............................................................................................................... 129 
ii.  RB and IPI-O constructs and Agrobacterium-mediated transformation ..................... 130 
iii.  Leaf Agroinfiltration on N. benthamiana and transgenic potatoes ............................. 131 
iv.  Molecular characterization of RB-SP and RB transgenic lines. ................................... 132 
v.  Protein interaction assays ............................................................................................ 133 
IV. Results ............................................................................................................................... 134 
i.  HR-induced assessment in N. benthamiana ................................................................ 134 
ii.  RB-SP/RB-transgenic potato lines ............................................................................... 135 
iii.  Coiled-Coil homodimerization disruption mediated by IPI-O4 .................................. 137 
V. Discussion ........................................................................................................................... 137 
LITERATURE CITED ............................................................................................................ 141 

 
Chapter 5 ..................................................................................................................................... 145 
GENERAL CONCLUSIONS ..................................................................................................... 145 

 

 
ix 

I. Improving pathogen-resistance in potato using genomic selection ..................................... 145 
II. Contributing to diploid potato breeding using targeted mutagenesis ................................. 146 
III. Understanding late blight resistance suppression .............................................................. 148 
IV. Final considerations ........................................................................................................... 149 
LITERATURE CITED ............................................................................................................ 151 

 

 
x 

 
 

 

LIST OF TABLES 

 
 

 

Table 2.1. Sequence of models. .................................................................................................... 52 
 
Table 2.2. Variance components estimates (posterior standard deviation) derived from BayesB 
model for late blight and common scab resistance by model. ....................................................... 57 
 
Table 2.3. Cross-validation correlations obtained with BRR and BayesB models by trait and 
model. ............................................................................................................................................ 60 
 
Table 2.4. Year cross-validation correlations obtained with BayesB model by trait and model. . 61 
 
Table S2.1. Late blight clonal lineages detected throughout seven years at the Michigan State 
University Clarksville Research Center (Clarksville, MI). ........................................................... 73 
 
Table S2.2. Variance components estimates (posterior standard deviation) derived from BRR 
model for late blight and common scab resistance by model. Phenotypic scores were 
standardized to unit variance; hence estimates can be interpreted as the proportion of variance 
explained by each component. ...................................................................................................... 74 
 
Table S2.3. Top ten ranked loci from the whole genome regression derived from BayesB model, 
according to their additive variance estimates for late blight. ....................................................... 75 
 
Table S2.4. Top ten ranked loci from the whole genome regression derived from BayesB model, 
according to their additive variance estimates for common scab. ................................................. 76 
 
Table S2.5. Year cross-validation correlations obtained with BRR model by trait and model. ... 77 
 
Table 3.1. S-RNase sequences from seven Solanaceae species used in this study. ...................... 90 
 
Table 3.2. DRH-195 and DRH-310-derived T0 and indel-based selected S-RNase KO lines with 
bi-allelic mutations. ..................................................................................................................... 101 
 
Table S3.1. Retrieved Illumina reads for genomic assembly of the DRH potato lines and 
assembly statistics. ...................................................................................................................... 112 
 
Table S3.2. Primers and sgRNAs used in this study. ................................................................. 113 
 
Table S3.3. Top TBLASTN hits of reported S-RNase proteins in the DM genome assembly. .. 114 
 
Table S3.4. Fruit set and seed count upon self-pollination in DRH-195 wild type and S-RNase-
derived KO lines. ......................................................................................................................... 115 
 

 

 
xi 

Table S3.5. Fruit set and seed count upon self-pollination in DRH-310 wild type and S-RNase-
derived KO lines. ......................................................................................................................... 117 
 
Table 4.1. Primers used in this study .......................................................................................... 132 
 
Table 4.2. Transformation efficiency for RB-transgenic lines. ................................................... 135 
 
 
  
 
 

 

 

 

 

 
xii 

LIST OF FIGURES 

 
 
 
Figure 1.1. Annual potato production from 1961 to 2017. ............................................................. 6 
 
Figure 1.2. Gametophytic self-incompatibility in Solanaceae. ....................................................... 9 
 
Figure 1.3. Conventional vs. genomic selection-based breeding in potato. ................................. 20 
 
Figure 2.1. (a) Boxplot of late blight scores (relative area under the disease progress curve- 
RAUDPC), and (b) bar plot for common scab scores (0-5 rating scale). ...................................... 55 
 
Figure 2.2. Principal component analysis of the Michigan State University’s potato breeding 
genotypes derived from 4,110 SNPs ............................................................................................. 55 
 
Figure 2.3. Heatmap of the genomic relationship matrix (GRM) from the Michigan State 
University’s potato breeding genotypes. ....................................................................................... 56 
 
Figure 2.4. Estimated SNP-variances derived from BayesB model using the additive model for 
late blight (a) and common scab (b). ............................................................................................. 59 
 
Figure S2.1. Linkage disequilibrium plots for the top ten ranked loci according to their 
contribution to the late blight genotypic variance. ........................................................................ 71 
 
Figure S2.2. Linkage disequilibrium plots for the top ten ranked loci according to their 
contribution to the common scab genotypic variance. .................................................................. 72 
 
Figure 3.1. S-RNase gene structure and allelic variants in diploid potato and related species. .... 95 
 
Figure 3.2. S-RNase gene mapping in diploid potato. .................................................................. 98 
 
Figure 3.3. CRISPR/Cas9 mediated mutagenesis in two SI diploid potato lines. ...................... 100 
 
Figure 3.4. S-RNase expression and knock-out phenotype in SI diploid potato lines. ............... 102 
 
Figure 3.5. Fruit formation in WT and T1 S-RNase KO diploid potatoes lines. ......................... 104 
 
Figure S3.1. S-RNase open reading-frames in T0 knock-out (KO) lines derived in two SI diploid 
potato lines. .................................................................................................................................. 111 
 
Figure 4.1. Induction of hypersensitive response (HR) in Nicotiana benthamiana. ................... 134 
 
Figure 4.2. Agroinfiltration of IPI-O effectors and RB-SP/RB insertion and expression in DRH-
S5 transgenic potato lines. ........................................................................................................... 136 
 

 

 
xiii 

Figure 4.3. Coiled-Coil (CC) domain dimerization assay using yeast-two-hybrid. ................... 137 

 

 
xiv 

ADVANCES IN GENOMIC STRATEGIES AND APPLICATIONS IN POTATO 

Chapter 1  

BREEDING 

 
I. Potato origin and domestication 

Potato (Solanum tuberosum L., Group Tuberosum) represents one of the most economically 

important species within the Solanaceae family along with tomato, eggplant, pepper, tobacco and 

petunia (Wu and Tanksley, 2010; Olmstead et al., 2008). The center of origin of the potato 

covers the Peruvian and Bolivian Andes where it evolved as a result of the hybridization of 

Solanum stenotonum complex with an unknown species (Grun, 1990). From there, the short-day 

sub-species S. tuberosum Group Andigena spread across the Andes, resulting later in the 

emergence of the long-day sub-species S. tuberosum Group Tuberosum (Hosaka, 2003; Hosaka 

and Hanneman, 1988). 

Potato domestication started 3400 calendar years before the present (cal BP) in the western 

Titicaca basin, contributing to the transition from nomadism to sedentarism of the first 

settlements in this region (Rumold and Aldenderfer, 2016). After the Spanish colonization, the 

potato was introduced to Europe in the 16th century (Hawkes and Francisco-Ortega, 1993). 

Later, S. tuberosum Group Tuberosum was introduced, replacing old varieties to face the 

damages caused by late blight, one of the most devastating diseases affecting potato (Ames and 

Spooner, 2008; Hosaka and Hanneman, 1988). From there, the potato spread across Europe, 

Africa, and Asia and later were re-introduced to North America by the British colonists in the 

17th century (Hawkes, 1992). 

Currently, potatoes are the third most important food crop after rice and wheat (Devaux et al., 

2014). In 2017, potatoes reached a global production of 388 million tons, positioning the United 

 

1 

States as the fifth largest producer with more than 20 million tons (FAOSTAT, 2019). As a 

staple food, potato plays an important role in human nutrition. Along with beans, potato is 

considered the cheapest source of fiber, minerals and vitamins necessary for child nutrition 

(Drewnowski and Rehm, 2013). It is also considered one of the main sources of carbohydrates 

(Kolasa, 1993). Besides its nutritional value, potato has high concentrations of phenolic 

compounds such as chlorogenic acid, opening a new market for potato’s commercialization as a 

functional food due to its potential anti-oxidative activity (Friedman, 1997; Vinson et al., 2012). 

 

II. Potato genetics and genomics-based resources 

Among the angiosperms, the Solanaceae represents one of the major families, with more than 

3000 species distributed worldwide (Wu and Tanksley, 2010). Within this family, Solanum is the 

biggest and most diverse genus, with more than 1250 species from shrubs to small trees (Weese 

and Bohs, 2007, Magoon et al., 1962). Solanum is classified into different sections, in which the 

Petota section is the most prominent group (Spooner et al., 1991). This section constitutes about 

100 tuber-bearing species, ranging from wild relatives to cultivated potato varities distributed 

from United States to Chile (Ovchinnikova et al., 2011). Its members are predominantly diploid 

(2x=2n=24), although tetraploid (2n=2x=48) and hexaploid (2n=6x=72) species are also present 

(Huamán and Spooner, 2002).  

Extensive genomic resources have been developed to decipher potato diversity and 

complexity. Genotyping platforms including 8K-20K Infinium Single Nucleotide Polymorphism 

(SNPs) arrays (Felcher et al., 2012b; Vos et al., 2015) and Genotyping-by-Sequencing (GBS) 

approaches have been widely used for potato germplasm characterization and allele variants 

discovery through Quantitative Trait Loci (QTL) mapping and Genome-Wide Association 

 

2 

(GWA) analysis (Bastien et al., 2018; Boudhrioua et al., 2017; Ellis et al., 2018; Endelman and 

Jansky, 2016; Manrique-Carpintero et al., 2015; Massa et al., 2015). 

Cultivated potato S. tuberosum is a heterozygous tetraploid species. Therefore, to reduce the 

complexity The Potato Genome Sequencing Consortium used a homozygous doubled 

monoploid, S. tuberosum Gp. Phureja line 1-3 516 R44 (DM) to generate a genome assembly 

(The Potato Genome Sequencing Consortium, 2011). This reference genome sequence was 

further improved by integrating physical and genetic maps from the heterozygous diploid lines: 

RH 89-039-16 S. tuberosum Gp. Tuberosum (RH); D84 (an S. tuberosum × S. chacoense hybrid) 

and DRH (DM × RH hybrid). A total of 674 Mb from the 727 Mb of the DM assembly were 

integrated, including 37,482 of the 39,031 predicted genes (Felcher et al., 2012a; Sharma et al., 

2013). To date, more than 200 potato accessions including wild and cultivated relatives from the 

Petota section have been sequenced (Hardigan et al., 2016; Li et al., 2018). In particular, 

sequencing data from species such as S. americanum, S. chacoense, S. commersonii, S. 

verrucosum among others (Aversano et al., 2015; Leisner et al., 2018; Lister et al., 2019; 

Paajanen et al., 2017), offer a new resource for mining disease-resistance genes and tuber-related 

traits. 

Whole-genome sequencing projects have contributed to elucidating the genetic variation 

landscape in diploid and tetraploid potatoes. The high degree of structural variation in potato, 

revealed by genomic and cytological analysis, significantly contributes to the high genomic 

diversity of this species (Hardigan et al., 2016; Iovene et al., 2013). In particular, potato 

heterozygosity and genetic load are driven by multiallelism associated with multiple 

polymorphisms in coding regions and Copy Number Variants (CNV), respectively (Pham et al., 

2017). Genomic analysis also demonstrated that potato domestication led to differential allele 

 

3 

expression and accumulation of deleterious mutations in tetraploid potatoes, this last 

compensated with the presence of heterozygous alleles, masking unfavorable mutations (Lian et 

al., 2019; Pham et al., 2017). 

Comprehensive gene expression atlases have been generated to understand the dynamics of 

key regulatory genes associated with physiological responses under different biological and 

environmental treatments in potato. Transcriptomic profiles from DM using up to 32 vegetative 

and reproductive tissues under different growth and stress conditions detected more than 22,000 

differential expressed genes (DEG) and co-expression gene modules associated to specific stress 

conditions (Massa et al., 2011, 2013). Additional transcriptomic profiling studies have uncovered 

key genes involved in metabolic pathways related to drought, salt tolerant and environmental-

stress adaptation responses (i.e., auxin response factors and nitrogen supplementation) using de 

novo or available expression repositories (Charfeddine et al., 2015a, 2015b; Gálvez et al., 2016; 

Gong et al., 2015; Song et al., 2019; Sprenger et al., 2016). 

Targeted and non-targeted metabolite profiles from various tissues and agronomic-related 

traits complement these transcriptomic atlases in potatoes. The chemical composition of potato 

leaves, tubers and resistance-related compounds have elucidated important metabolic pathways 

associated with the synthesis of carbohydrates, glycoalkaloid, vitamins and bioactive compounds 

such as hydroxycinnamic acid (Chaparro et al., 2018; Mariot et al., 2016; Piñeros-Niño et al., 

2017; Rodríguez-Pérez et al., 2018; Tomita et al., 2017; Yogendra et al., 2015). Additional 

transcriptomic and metabolomic repositories are also available for autotetraploids derived from 

the wild relatives S. commersonii and S. bulbocastanum, representing a new resource for 

polyploidization and gene discovery studies (Fasano et al., 2016). 

 

4 

The aforementioned approaches have enabled researchers to identify, validate and 

reconstruct historical introgressions of adaptation-related traits such as plant maturity. For 

instance, association analyses and homologous searching of S. tuberosum Gp Tuberosum RH 89-

039-16 (RH) genomic sequences revealed a major QTL controlled by the S. tuberosum Cycling 

Dof Factor (StCDF gene), located in Chromosome 5 (Collins et al., 1999; Kloosterman et al., 

2013). Functional analysis confirmed the role of StCDF in plant maturity through 

complementation assays using the short-day S. tuberosum Group Andigenum (Kloosterman et 

al., 2013). Further investigations revealed the role of StCDF in the short-day to long-day 

transition through the introgression of this gene from diploid landraces into cultivated tetraploid 

potatoes (Hardigan et al., 2017). Together, these genomic resources offer a valuable tool for gene 

mining, evolutionary studies, germplasm characterization and potato breeding. 

 

III. Vision for diploid potato breeding – Advantages and Challenges 

Conventional breeding and genetic analyses are challenging in cultivated potato due to its 

tetraploidy, heterozygosity and vegetative mode of propagation. Re-inventing potato as a diploid 

inbred/F1 hybrid variety (2n=2x=24) would allow the application of efficient breeding methods 

(Jansky et al., 2016). For instance, the development of inbred diploid potatoes not only would 

accelerate the generation of new varieties with favorable allele combinations targeting yield, 

tuber quality, and disease resistance traits but will also aid in understanding the genetics 

underlying these traits. Diploid based approaches will enable increases in genetic gains per 

breeding cycle (Jansky et al., 2016). Compared with major staple crops such as maize and rice, in 

which heterosis has contributed significantly to their increase in global production (Fu et al., 

2014), production increases have been non-significant in potato under the current breeding 

 

5 

system during the past decades (Figure 1.1). Therefore, diploid potato breeding could achieve 

comparable gains by exploiting heterosis supported with new breeding strategies. 

Improvement of cultivated tetraploid potatoes also relies on the discovery and introgression 

of genes from wild species, especially for traits related to biotic and abiotic stresses. However, 

the introgression of critical dominant alleles in a triplex or quadruplex allelic configuration can 

take up to 15 years in a tetraploid breeding scheme (Mendoza et al., 1996). Moreover, vegetative 

propagation favors disease transmission (i.e., viruses), which requires the production of disease-

free tubers increasing production costs (Simmonds, 1997). 

900

800

700

600

500

400

300

200

100

0

 

s
n
o
T
n
o
i
l
l
i

M

 

 

1
6
9
1

3
6
9
1

5
6
9
1

7
6
9
1

9
6
9
1

1
7
9
1

3
7
9
1

5
7
9
1

7
7
9
1

9
7
9
1

1
8
9
1

3
8
9
1

5
8
9
1

7
8
9
1

9
8
9
1

1
9
9
1

3
9
9
1

5
9
9
1

7
9
9
1

9
9
9
1

1
0
0
2

3
0
0
2

5
0
0
2

7
0
0
2

9
0
0
2

1
1
0
2

3
1
0
2

5
1
0
2

7
1
0
2

Potato

Rice

Wheat

Figure 1.1. Annual potato production from 1961 to 2017. No significant increases in potato 

production has been observed when compared with the rice and wheat. Source: FAOSTAT 

(March 8, 2019). 

 

 

6 

Moreover, diploid potatoes will enable the commercialization of this crop through true seed 

instead of seed tubers for propagation, reducing transportation costs as well as disease 

transmission of bacterial and oomycete pathogens (Jansky et al., 2016). Additionally, genomics 

tools developed in other diploid species could now be accessible for potato. Allele dosage 

estimation, quantitative trait loci detection, breeding value estimation among others will be more 

feasible and accurate under a diploid system (Jansky et al., 2016). 

One of the major challenges for future diploid breeding programs is male sterility. This 

phenomena occurs when Solanum relatives are used as males and crossed to S. tuberosum-

derived dihaploids lines (Carroll, 1975; Jansky et al., 2016; Jansky and Peloquin, 2006), 

representing a limiting factor for diploid hybrid production. Linkage drag represents another 

major drawback for introgressive hybridization when using wild-potato relatives as donors 

(Gaiero et al., 2018). For instance, introgression of resistance-related genes into cultivated 

diploid germplasm could lead to fixation of undesirable traits such as high tuber glycoalkaloid 

content from donors such as S. chacoense. However, the gametophytic self-incompatibility (SI) 

represents one of the most significant barriers to diploid breeding, preventing the ability to 

generate diploid homozygous lines. 

 

i. Self-incompatibility in diploid potatoes 

In Solanaceae, a single multiallelic S-locus governs self-pollen rejection under the gametophytic 

SI system. This locus is composed of tightly linked genes, SLF (S-locus F-box) and S-RNase (S-

locus RNase), expressed in the pollen and the style, respectively (McClure et al., 1989; 

Takayama and Isogai, 2005). The S-RNase protein produces cytotoxic effects that inhibit the 

elongation of self-pollen tubes via degradation of RNA from the pollen whereas SLF functions as 

 

7 

a component of a detoxification complex that mediates ubiquitination of non-self S-RNase 

proteins leading to degradation via the proteasome pathway (Kubo et al., 2015; Sijacic et al., 

2004). Hence, when self-pollination occurs in self-incompatible individuals, SLF does not 

recognize its native S-RNase, and consequently, pollen tube growth in the style is inhibited due 

to the ribonuclease activity of the S-RNase (Hua et al., 2008). 

Most species under the GSI system contain multiple SLFs to recognize a broader set of non-

self S-RNases (Figure 1.2A) (Fujii et al., 2016; Kubo et al., 2010). This collaborative non-self-

recognition system is under the control of the S-haplotype-specific F-box brothers (SFBB) as 

result of SLF inter-haplotype genetic exchanges (Kubo et al., 2015; Sassa et al., 2007). SFBB 

proteins provide an evolutionary advantage for outcrossing, increasing the number of potential 

mating partners, but also contributing to a self-compatibility transition recognizing self-pollen 

(Fujii et al., 2016; Kubo et al., 2010, 2015; Sassa et al., 2007).  

Although GSI gene determinants confer specificity for pollen rejection, modifier genes are 

also required to modulate the SI response (Goldraij et al., 2006; McClure et al., 1999; O’Brien et 

al., 2002). For instance, pollen determinant modifiers such as the S-RNase binding protein 1 

(SBP1), a ring domain protein, form a complex with SLF, interacting with S-RNase conserved 

regions between hypervariable domains, leading to non-self S-RNase degradation (O’Brien et al., 

2004; Sassa et al., 2007). Similarly, stylar factors such as the 120-KD arabinogalactan (120K) 

and Kunitz-type proteinase inhibitors (NaStEP) play an essential role in the pistil endomembrane 

protein trafficking and S-RNase toxicity (Figure 1.2B) (Goldraij et al., 2006; Jiménez-Durán et 

al., 2013; Lee et al., 2009). 

Two different strategies have been used to develop diploid-self-compatible (SC) potato lines. 

The inbred line M6, generated from the wild tuber-bearing species, S. chacoense (Jansky et al., 

 

8 

2014) harbors a dominant allele of the S-locus inhibitor (Sli) that inactivates the gametophytic SI 

system (Hosaka and Hanneman, 1998) leading to self-compatibility. However, introgression of 

Sli into other germplasm is time-consuming and could lead to linkage drag and fixation of 

undesirable traits such as high tuber glycoalkaloid content from the donor S. chacoense. 

Similarly, a targeted mutagenesis approach has been used to target the S-RNase gene, 

successfully generating self-compatible potato lines (Ye et al., 2018). 

Figure 1.2. Gametophytic self-incompatibility in Solanaceae. (a) The S-locus is composed of 

tightly linked genes, S-RNase (S-locus RNase) and multiple SLFs (S-locus F-box proteins), 

 

9 

which are expressed in the style and pollen, respectively. (b) After pollination, S-RNase is 

imported into pollen tubes along with proteins such as 120K, which is required for their ability to 

reject incompatible pollen. In self-compatible reactions, the N terminal F-box domain from SLF 

recognize components of the SCF E3 ubiquitin ligase complexes. Likewise, SBP1 interact with 

the S-RNase N-terminal region and E3 ubiquitin ligases forming a complex with SLF-SCF, 

leading to the degradation of non-self S-RNases. In self-incompatible reactions, the SLF-SCF 

complex is not able to recognize its own S-RNase leading to the degradation of RNA involved in 

self-pollen tube development by the ribonuclease activity of S-RNase. 

 

IV. Current production constraints in cultivated tetraploid potatoes 

Despite its significant economic importance, potato faces production losses, with 1,331 tons in 

the United States alone (National Potato Council, 2016). While abiotic factors such as heat and 

drought can be one of the major limiting components in potato production, affecting the growing 

cycle and reducing tuber yield and quality, biotic factors have the most devastating impacts over 

this crop (George et al., 2017; Kromann et al., 2014). Pests and pathogens have a severe effect on 

sustainable potato production, generating production losses in field and storage (Kromann et al., 

2014). The most recent data 2001-2003 (though it is now nearly 20 years old) shows that 

pathogens caused 14.5% of global potato production losses, followed by pests with 10.9 %, 

weeds with 8.3 and viruses with 6.6%, and it is estimated that without pest control total 

production losses could reach up to 75% (Oerke, 2006). Pathogens such as late blight and 

Alternaria are the most devastating and widely distributed constraints on potato, along with the 

viruses like potato leafroll luteovirus and potato potyvirus Y. Animal pests like potato cyst 

nematodes and the Colorado beetle are significant as well (Oerke, 2006). Among potato 

 

10 

pathogens, common scab and late blight are two primary production constraints faced by potato 

growers, causing up to 100% of losses, and reducing the quality and marketability of the tuber. 

Although pathogen infection can be controlled by using protectants or systematic fungicides 

in potato, they can be ineffective if environmental conditions favor pathogen dispersion or 

because of the emergence of fungicide-resistant genotypes (Nowicki et al., 2011; Pomerantz et 

al., 2014). The most effective way to control the incidence of common scab and late blight in 

potato is through the generation of resistant varieties (Ahn and Park, 2013) and therefore this 

research seeks to contribute to the generation of resistant potato varieties using different genomic 

approaches, contributing to the reduction of the potato breeding cycle. 

 

i. Common scab (Streptomyces scabies Thaxter) 

Common scab is a soil-borne pathogen that reduces potato quality and marketability by 

causing superficial lesions on the tuber surface (Dees and Wanner, 2012). This pathogen, 

previously considered a fungus given its filamentous morphology, is a gram-positive bacteria 

which produces spores that allow it to survive on different substrates (Braun et al., 2017). 

Common scab penetrates young tubers through lenticels, spreading rapidly in growing plants 

cells, facilitated by the phytotoxin thaxtomin produced by this pathogen (Braun et al., 2017). 

This pathogen has a worldwide distribution, and due to its saprophytic nature, common scab can 

survive in winter, thus becoming a permanent source of inoculum for the next planting seasons, 

causing losses up to $100/Ha (Wanner and Kirk, 2015). Susceptibility to common scab is 

dependent upon genotype, growing season and environmental conditions (Wanner, 2006; 

Wanner and Kirk, 2015), having a negative impact mainly in underground tissues in 

development, such as stolons and tubers. 

 

11 

ii. Late blight [Phytophthora infestans (Mont.) de Bary] 

Late blight is known for being the causal agent of the Irish potato famine in the 19th century, 

leading to over a million human deaths (Grada, 2004). infestans a hemibiotrophic oomycete, 

originated in central Mexico and later spread all over the world (Grünwald and Flier, 2005). At 

the first stage of infection, late blight forms an asexual multinucleate sporangium or 

mononucleate zoospores produced by cleaved sporangia, entering into the host via stomata, 

wounds or appressorium formation in vegetative tissues (Leesutthiphonchai et al., 2018). Late 

blight also forms sexual spores known as oospores, that require two mating types (A1 and A2). 

Through sexual recombination, this mating system increases pathogen diversity when both 

mating types are present in a population, favoring late blight adaptation to extreme 

environmental conditions (Smart and Fry, 2001). Consequently, due to its ability to use both 

sexual and asexual reproduction, using wind and rain as a means of dispersal, P. infestans kills 

an entire plant within 7 to 10 days, and is responsible for annual losses of 16% to global potato 

production (Haverkort et al., 2009). 

 

iii. Late blight-potato arms race 

Plants lack an adaptive immune system, and therefore disease resistance relies on the 

recognition of molecular signatures associated with pathogens. Plant cells display a coordinated 

system to recognize threats through extramembrane receptors coupled with intramembrane 

kinases (Couto and Zipfel, 2016). These Pattern Recognition Receptors (PRRs) perceive 

pathogen-associated molecular patterns (PAMPs) or host-derived damage-associated molecular 

patterns (DAMPs), inducing a signaling cascade that activates the expression of defense-related 

genes, a process known as PRR-triggered immunity (PTI) (Couto and Zipfel, 2016; Jones and 

 

12 

Dangl, 2006). 

However, fast-evolving pathogens promptly respond to PTI, translocating virulent effectors 

into the host cell, mostly via the Type Three Secretion System (TTSS), suppressing PTI. In turn, 

plants display an arsenal of intracellular receptors encoded by R genes, inducing a second layer 

of resistance response known as Effector-Triggered Immunity (ETI) upon sensing or recognition 

of pathogen effectors (Couto and Zipfel, 2016; Jones and Dangl, 2006). These multidomain 

protein receptors are composed of a Nucleotide-binding domain Leucine-rich Repeat (NLR) core 

with either a Coiled-Coil (CC) or a Toll Interleukin-1-like receptor (TIR) domain at the N 

terminal region (Sukarta et al., 2016). This host-pathogen arms race has shaped the genetic 

architecture of plant disease resistance. Previously, it was widely accepted that plant resistance 

responses were associated with a gene-by-gene or R-gene-mediated (Flor, 1971). However, 

current models suggest that plant pathogen resistance results from a dynamically co-evolving 

process involving resistance breakdown and recovery, known as the zig-zag model (Cook et al., 

2015; Jones and Dangl, 2006). Understanding plant-pathogen interaction may provide a 

mechanism to counteract host resistance suppression in potato by modifying effector targeted 

sites, enhancing R gene-mediated resistance. 

 

iv. Genetic architecture of late blight resistance in potatoes 

The outcrossing nature of potato has enabled the transfer of genetic pools from wild species 

through hybridization, despite the partial sterility and different ploidy levels. Due to its closely 

phylogenetic distance, wild tuber bearing-potato have been the primary resistant genes source in 

cultivated potato (Hawkes, 1958). Thus, for late blight, resistance to P. infestans (Rpi) genes 

have been discovered using high-resolution maps and positional cloning in Solanum section 

 

13 

Petota species such as S. venturii (Rpi-vnt1.1 and Rpi-vnt1.3), S. pinnatisectum (Rpi1) and S. 

phureja (Rpi-phu1). These particular genes are distributed across four potato chromosomes (5, 

6,7, and 9) with a common CC-NB-LRR gene architecture (Kuhl et al., 2001; Park et al., 2005a; 

Pel et al., 2009; Vossen et al., 2005). Similar resistant genes/loci such as R8-R9 and Rpi-abpt 

have been isolated from S. bulbocastanum, S. demissum and a quadruple hybrid (S. acaule, S. 

bulbocastanum, S. phureja and S. tuberosum), resulting in high resistance levels against different 

late blight strains (Park et al., 2005b). Additionally, different resistant genes/loci have been 

identified within S. tuberosum cultivars, including R1, R3, R6 and R7 (El-Kharbotly et al., 1996). 

R1, a NLR protein, is located in a hotspot region on Chromosome 5, clustered with genes that 

confer late blight resistance in potatoes (Ballvora et al., 2002). 

Recently, the Resistance gene enrichment Sequencing (RenSeq) technology has enabled 

the identification of 331 unidentified R genes in potato, distributed across the potato genome, 

using NBL-targeting baits and Illumina sequencing (Jupe et al., 2013). This system has allowed 

identifying new late blight resistance genes such as Rpi-amr3it from S. americanum using single-

molecule real-time (SMRT) sequencing, coupled with RenSeq (SMRT RenSeq) (Witek et al., 

2016). 

 

v. RB-mediated resistance in potato 

The RB gene or rpi-blb1 (resistant protein to P. infestans from S. bulbocastanum) is a member of 

a gene family with four genes located on Chromosome 8 of S. bulbocastanum (Song et al., 

2003). This gene confers broad-spectrum disease resistance against late blight and has been 

successfully incorporated into different potato cultivars via Agrobacterium-mediated 

transformation (Halterman et al., 2008). Although it has been observed to result in partial 

 

14 

resistance in field, RB-carrying potatoes present a foliar resistant increase in growth chamber and 

greenhouse compared to wild type genotypes, without significant effects on tuber yield 

(Halterman et al., 2008). 

The RB gene encodes a protein of 970 aa with typical R gene architecture, containing an N-

terminal CC domain along with an NB (with a p-loop and two kinase motifs) and LRR (with 21 

LRR motifs mostly imperfect) domains (Song et al., 2003). After infection, the avirulent protein 

IPI-OI, a class I effector from P. infestans, is translocated into the host cytoplasm, interacting 

with the CC domain from RB triggering a host defense responses (Chen et al., 2012).  

Under normal conditions, RB remains in a resting state. After pathogen infection, RB 

proteins dimerize through CC domains mediated by IPI-O1 recognition. However, late blight 

strains carrying the class III variant effectors IPI-O4 suppress IPI-O1-mediated resistance by 

direct competition for the CC binding site, avoiding the CC dimerization (Chen et al., 2012; 

Chen and Halterman, 2017; Halterman et al., 2010). IPI-O4-carrying late blight strains cause 

more disease than those without it. However, this response is cultivar dependent indicating a 

possible presence of extra components acting against late blight infection, reducing the resistance 

response (Chen et al., 2012). The CC domain from the S. pinnatisectum RB homologous does not 

exhibit any interaction with IPI-O4, suggesting that this species may escape the IP-O4-mediated 

resistance suppression (Chen et al., 2012). Therefore, engineering the RB-protein may contribute 

to the generation of a durable late blight resistance response in RB-carrying potato. To achieve 

this goal, it is imperative to validate the IPI-O4 resistance suppression mediated in potato. 

Leveraging the molecular interactions between late blight and potato will enable the generation 

of durable late blight resistance, adopting strategies like CC domain swapping between the RB 

genes from S. pinnatisectum and S. bulbocastanum, as proposed by Chen et al. (2012). 

 

15 

V. Genomic approaches apply to potato breeding 

Breeding for resistant varieties via phenotypic selection can take up to 15 years, which is time-

consuming and ineffective against fast-evolving pathogens in potatoes (Lozano et al., 2012; The 

Potato Genome Sequencing Consortium, 2011). The use of statistical models in modern plant 

breeding has revealed new resistance gene sources, dissected the disease/resistance architecture 

and uncovered the mechanisms in which plants respond to pathogen attack (Boyd et al., 2013). 

Moreover, the capture and prediction of genetic variation for disease resistance using prediction 

models have contributed to maximizing the genetic gains while decreasing the breeding cycle in 

staple crops such as maize and wheat (Poland and Rutkoski, 2016). In potato, these strategies can 

support conventional breeding schemes, increasing the genetic gains and accelerating the 

generation of resistant potato varieties. The achievement of this goal must rely on the 

implementation of appropriate statistical models that consider the genetic complexity and allele 

dosage of this crop. 

 

i. Whole genome association (GWA) analysis in potato 

With the advent of high throughout next generation sequencing technologies (HT-NGS), it is 

now feasible to identify variants associated with disease resistance with relatively high resolution 

using GWA analysis. Unlike bi-parental QTL (quantitative trait locus) mapping which is 

restricted to the genetic variation coming from two parental lines, GWA is able to detect 

historical recombination events in germplasm collections, exploiting natural diversity, increasing 

mapping resolution and reducing research time (Yu and Buckler, 2006). GWA analysis relies on 

the linkage disequilibrium (LD) between a causative gene variant and candidate markers used for 

genotyping. Besides LD, GWA ability for QTL detection also depends on the population size, 

 

16 

marker density, relatedness and the control of non-genetic-related factors such as experimental 

design or field replicates (Lipka et al., 2015). 

Confounding effects caused by population stratification can lead to false-positive discoveries 

decreasing the power of detection in association analysis (Hoffman, 2013). This problem was 

initially addressed through General Linear Models (GLM). Later, statistical methods such as 

principal component analysis (PCA), Bayesian model-based approaches and discriminant 

analysis of principal components (DAPC) were used to fit the hidden ancestry in a given 

population using Mixed Linear Models (MLM) (Corander and Tang, 2007; Hoffman, 2013; 

Lipka et al., 2015). These models also include random effects using variance-covariance 

matrixes of the individuals (kinship matrix) and have been widely used for QTLs detection in 

complex traits for most important crops (Zhang et al., 2010). Recently, Efficient Mixed-Model 

Association (EMMA) and Genome-wide Efficient Mixed-Model Association (GEMMA) 

algorithms have been used to address large marker and sample numbers, increasing the 

efficiency and computational speed in GWA analyses (Zhang et al., 2010; Zhou and Stephens, 

2012). Complementary approaches use compression methods such as compress-MLM and P3D 

decreasing significantly computational time, reducing the sample size, by clustering similar 

individuals, and estimating the population structure (Zhang et al., 2010). To address population 

structure effect, multi-parental populations capturing historical and recent recombination events 

have been generated. Nested Association Mapping (NAM), Multi-parent Advance Generation 

Intercrosses (MAGIC) and Recombinant Inbred Advanced Intercross Lines (RIAILs) combine 

the advantages of GWAS and bi-parental populations, producing new chromosomal re-

arrangements and increasing the genotypic diversity  (Bandillo et al., 2013; Huang et al., 2015; 

Lipka et al., 2015; Rockman and Kruglyak, 2008). 

 

17 

In potato, association analyses have been conducted using linear and non-parametric (Mann–

Whitney U test) models, for monogenic and complex traits such us pathogen/pest resistance, and 

tuber yield and quality, respectively (Rosyara et al., 2016; Schönhals et al., 2016). Though most 

of these studies have assessed allele dosage by using band intensities with AFLP (amplified 

fragment length polymorphism), overlapping peaks in electropherograms with SSRs (simple 

sequence repeats) and sequence variations with SNPs (single nucleotide polymorphisms) 

(Achenbach et al., 2009; D’hoop et al., 2014; D’Hoop et al., 2008; Gebhardt et al., 2004; 

Malosetti et al., 2007), considering both dominant and additive effects by the detection power of 

only an additive model, as a common practice in GWA (Bush and Moore, 2012), leading to 

biases in the final outcome. However, to date, no GWA analyses have been conducted to unravel 

the genetic architecture of disease resistance in potato, supported by high throughput genotyping 

technologies and appropriate statistical models representing the complex nature of potato. 

 

ii. Genomic selection (GS) in potato 

QTL mapping studies have contributed widely to the identification of genomic landscapes used 

for oligogenic trait selection through Marker Assisted Selection (MAS) to accelerate the 

generation of new cultivars. This molecular breeding approach depends on the reliability and 

tight linkage between the molecular marker and the causal gene of a particular trait (Collard and 

Mackill, 2008). GWA studies have reasonably good power to detect large effect-QTL in 

complex traits. However, GWA usually lacks power to detect associations for small effect 

variants. This can lead to high rates of missing heritability and low prediction accuracy (Desta 

and Ortiz, 2014).   

 

18 

To overcome this problem Meuwissen and Hayes (2001) proposed to use Whole-Genome 

Regression (WGR) methods for prediction of breeding values. This approach, also referred to as 

GS, has proved to be effective for prediction of breeding values in plant and animal breeding 

populations, such as maize and cattle, respectively (Beyene et al., 2015; Hayes et al., 2009; 

Zhang et al., 2014). A WGR model uses large number of variants (e.g., SNPs) distributed over 

the whole genome to predict breeding values for individuals with only genotypic information, 

using a trained model calculated from related individuals with both, genotypic and phenotypic 

information (Heffner et al., 2009). WGR can be extended by including SNPs detected in GWA 

analyses fitting them as fixed effects, increasing prediction accuracies (Spindel et al., 2016). 

In potato, estimated breeding values (EBV) have been generated using best linear unbiased 

prediction (BLUP) methods and pedigree information. This method showed an improvement in 

progeny performance when used for traits with low heritability such as yield and plant maturity 

(Slater et al., 2014). Likewise, potato breeding programs have widely used MAS to select lines 

with desirable traits such as starch content and virus Y resistance, respectively (Freyre and 

Douches, 1994; Hämäläinen et al., 1997). However, these approaches are not able to capture 

total genetic variances since they are restricted only to the variance explained by major QTLs 

(Heffner et al., 2009). To address this problem, GS has been implemented in potato for yield, 

tuber quality and biotic/abiotic-related traits (Slater et al., 2016). Since GS requires a high 

marker density distributed across the genome, the Infinium 8303 Potato Array containing a set of 

SNP markers distributed across the 12 potato chromosomes (Felcher et al., 2012a) has been used 

for this purpose. More recently, a greater SNP number derived from Genotyping By Sequencing 

(GBS) and Diversity Array Technology (DArTseq) silicoDArT have been used to increase 

genetic gains using GS, obtaining high cross-validation accuracies (0.25-0.7) for complex traits 

 

19 

such a yield (Habyarimana et al., 2017; Sverrisdóttir et al., 2017). Overall, these studies 

demonstrate the feasibility of GS implementation in potato breeding programs as a contribution 

to the reduction of the breeding cycles in this crop (Figure 1.3). In particular, GS can contribute 

to improving pathogen resistance in potato cultivars, promptly responding toward fast-evolving 

pathogens such as late blight. 

Figure 1.3. Conventional vs. genomic selection-based breeding in potato. Traditional breeding 

of potato relies on the generation of hundreds of crosses which progeny is evaluated and selected 

in the field up to 15 years to finally generated one or two new potato varieties. Genomic 

selection will reduce the breeding cycle up to seven years by selected superior lines based on 

 

their genetic merit using whole-genome regressions. 

 

iii. Genome editing: An efficient method to target agronomic-related traits 

GS has accelerated plant breeding, reducing the generation time of new cultivars by selecting 

individuals based on their genetic value. However, unlike GS, targeted mutagenesis (TM) 

represents a fast alternative to obtaining improved varieties by directly altering the expression of 

 

20 

genes controlling traits of interest. Genome editing technologies arise as an alternative tool for a 

precise and efficient targeted mutagenesis, since unlike conventional transformation, genome 

editing avoids the modification of regions other than the target sites (Bortesi and Fischer, 2015). 

Using double strand breaks (DSB), engineered endonucleases cuts the DNA at the target site, 

triggering the response of endogenous cell repair mechanisms. Through homology directed 

repair (HDR) or non-homologous end joining (NHEJ) repairs, deletions (knockout-KO) or 

insertions (knockin-KI) are achieved, altering the expression of a particular gene (Bortesi and 

Fischer, 2015; Pellagatti et al., 2015). 

 

a. Targeted mutagenesis technologies 

Initially, TM was achieved through native engineered homing endonucleases (HEs) targeting 

genomic regions spanning 12-40 pb, and generating DSB or single strand-breaks that activate 

either NHEJ or HDR (Hafez and Hausner, 2012). Homing endonucleases have been used to 

create gene KO/KI on immortalized human cell lines and plant genes involved in disease and 

pest control (Hafez and Hausner, 2012). However, this technology was limited to naturally 

occurring enzymes, reducing the number of possible target sites (Belfort and Bonocora, 2014). 

To address this problem, customized sequence-specific endonucleases such as Zinc Finger (ZF) 

and Transcriptions Activator-Like Effector (TALE) nucleases emerged to disrupt or correct 

predeterminate genomic regions (Bogdanove and Voytas, 2011; Urnov et al., 2010). Zinc Finger 

nucleases couples the DNA-binding domain from zinc finger transcription factors with the 

nuclease domain from the restriction enzyme FokI (Urnov et al., 2010). This TM approach uses 

Cys2 His2 fingers in tandem arrays that recognizes 3 bp each, increasing target specificity when 

used as a heterodimer, binding up to 12 bp of the intended genomic regions on each direction 

 

21 

(Miller et al., 2007). In contrast, TALENs recognize single contiguous base-pairs using tandem 

polymorphic amino-acid repeats (Bogdanove and Voytas, 2011). This technology relies on the 

DNA-binding ability of the TALE protein derived from plant pathogen effectors coupled with 

FokI nucleases as ZFN. Transcriptions Activator-Like Effector nucleases use an amino-

acid/DNA-type code to design effector proteins, whose specific amino-acid order recognizes an 

intended DNA sequence (Joung and Sander, 2013). This feature confers limitless targeting 

ranges unlike ZF nucleases which lack specific zinc finger combinations for some specific 

nucleotides triplets (Joung and Sander, 2013). 

Recently, a more precise, amenable and economic technology has become the preferred 

approach within the scientific community for gene targeting and genome edition. This 

technology known as Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR)-

associated Cas 9 protein (Cas9) system has been widely used to generate gene KO/KI of 

candidate genes related to agronomic traits in important crops (Jaganathan et al., 2018). The 

Cas9 target-DNA recognition is mediated by a single guide-RNA (sgRNA) bearing a 20 bp 

target-site complementary to a region adjacent to a protospacer-adjacent motif (PAM), 5′-NGG-

3’, resulting in the generation of a DSB (Doudna and Charpentier, 2014; Pellagatti et al., 2015; 

Sternberg et al., 2014) which can be leveraged to generate DSB of target genes. 

CRISPR/Cas9 has extended its range of genomic targets and applicability by altering the 

Cas9 structure and creating new protein complexes, enabled it not only to address functional but 

structural analyses. For instance, catalytically inactive Cas9 (dCas9) proteins have been used to 

turn on/off genes to investigate gene function and localization when fused with proteins such as 

green fluorescent proteins (Doudna and Charpentier, 2014). Likewise, specific point mutations 

can be generated using cytidine or adenosine deaminases coupled with dCas9, converting C/G to 

 

22 

A/T and A/T to C/G bases, minimizing the generation of defective DBS-related products (i.e., 

translocations) (Rees and Liu, 2018). The ability of the Cas protein to recognize a wide range of 

target DNA and RNA has also been expanded. Cas variants such as Cas12a and Cas13b are 

capable of generating single-stranded DNA cleavage and gene knockdown, respectively, 

expanding the range of application of this technology to study essential genes and non-coding 

RNAs (Chen et al., 2018; Cox et al., 2017). Similarly, PAM sequence specificities have been 

broadened using pre-existing (SpCas9-NG) or evolved (xCas9) Cas versions (Hu et al., 2018; 

Nishimasu et al., 2018) expanding targetable genome loci. In contrast, anti-CRISPR associated 

(Aca) protein families, have been identified and studied for their ability to inhibit the CRISPR 

RNA (crRNA) binding, Cas9 dimerization, and conformational changes for cleavage activity 

(Pawluk et al., 2017). Anti-CRISPRs could be used as a safety measure for gene drive-derived 

technologies and contribute to the reduction of off-target effects by controlling Cas expression 

(Pawluk et al., 2017). 

 

b. Genome editing in potato 

Targeted mutagenesis has been successfully implemented in potato mainly targeting herbicide 

resistance and tuber quality-related traits. Transient and stable transfections assays have been 

conducted in potato protoplast and callus using TALENS, targeting the Acetolactate Synthase 

gene (ALS) and Vacuolar Invertase gene (Vlnv) (Butler et al., 2015; Clasen et al., 2016; Nicolia 

et al., 2015). Likewise, CRISPR-Cas9 technologies have been used in potato to eliminate the 

production of steroidal glycoalkaloids and Aux/IAA proteins by knocking-out the St16DOX and 

StIAA2 genes respectively (Nakayasu et al., 2018; Wang et al., 2015). CRISPR-Cas9 

ribonucleoproteins (RNPs) have also been delivered into potato protoplasts, targeting the 

 

23 

Granule Bound Starch Synthase (GBSS) encoded gene to prevent Cas9 integration into the 

potato genome (Andersson et al., 2018). 

Base editing has also been achieved in potato. For instance, specific point mutations targeting 

the GBSS and ALS genes have been generated using base editing in protoplasts, coupling the 

human A3A deaminase with a Cas9 nickase, increasing the deamination window to 17 

nucleotides compared to the five nucleotides observed in other systems (Zong et al., 2018). 

However, the complex nature of potato limits TM for targeting alleles on each potato 

homologous chromosomes. To address this limitation, an enhancer derived from the OsMac3 

mRNA 5′-untranslated region (dMac3) has been introduced into a CRISPR/Cas9 cassette to 

increase chromosome-specific allele mutagenesis. dMac3 boosts Cas9 synthesis, increasing the 

ratio of GBSS KOs containing four mutant alleles (Kusano et al., 2018). 

 

VI. Project objectives 

This project aims to integrate genomic selection and genome editing strategies to accelerate 

potato breeding, contributing to reduce the breeding cycle. A strategy to understand the 

molecular interaction between potato and late blight is also presented. First, in Chapter 2 the 

generation of allele dosage models in tetraploid potato is presented under the title “Genomic 

selection for late blight and common scab resistance in tetraploid potato (Solanum tuberosum).” 

This research seeks primarily to develop an allele dosage model for polyploids that accounts for 

additive and dominance effects and to use that model to detect loci associated with late blight 

and common scab resistance in tetraploid potatoes. As a second specific aim, the expectation is 

to integrate the models developed into models for genomic prediction and to evaluate the ability 

of these models to predict resistance to late blight and Scab in a tetraploid potato population. 

 

24 

Second, in Chapter 3, a genome editing approach is implemented to breakdown the S-RNase-

based self incompatibly in potato, under the title “Overcoming self-incompatibility in diploid 

potato using CRISPR-Cas9”. The purpose of this chapter was first to identify S-RNase allelic 

variants in two self-incompatible diploid potatoes. As a second specific aim, it is expected to 

generate S-RNase KO lines using the CRISPR-Cas9 system in self-incompatible diploid 

potatoes. Finally, the third specific aim involves characterizing the gene-edited T0 events and 

their phenotypic evaluation for self-compatibility. 

Third, in Chapter 4, in vitro and in vivo approaches were used to confirm the interaction 

between RB CC domains and IPI-O effectors from late blight, under the title “Validation of the 

RB mediated-resistance suppression by the class III effector IPI-O4 from late blight”. First, using 

RB and IPI-O co-infiltration assays on N. benthamiana and RB-transgenic potato lines, it is 

expected to confirm previously reported in vitro and in vivo observations for this pathosystem. 

Finally, a second specific aim seeks to validate the IP-O4 suppression effect over CC-self 

interaction using a yeast three-hybrid approach. Finally, the concluding Chapter 5 is presented, 

summarizing the main findings, drawbacks and future directions of each research chapter. 

 

 

 

 

 

25 

LITERATURE CITED

 

26 

LITERATURE CITED 

 

 
 
Achenbach, U., Paulo, J., Ilarionova, E., Lübeck, J., Strahwald, J., Tacke, E., et al. (2009). Using 

SNP markers to dissect linkage disequilibrium at a major quantitative trait locus for 
resistance to the potato cyst nematode Globodera pallida on potato chromosome v. 
Theoretical and Applied Genetics 118, 619–629. 

 
Ahn, Y. K., and Park, T.-H. (2013). Resistance to common scab developed by somatic hybrids 

between Solanum brevidens and Solanum tuberosum. Acta Agriculturae Scandinavica, 
Section B - Soil & Plant Science 63, 595–603. 

 
Ames, M., and Spooner, D. M. (2008). DNA from herbarium specimens settles a controversy 

about origins of the European potato 1. 95, 252–257. 

 
Andersson, M., Turesson, H., Olsson, N., Fält, A.-S., Ohlsson, P., Gonzalez, M. N., et al. (2018). 

Genome editing in potato via CRISPR-Cas9 ribonucleoprotein delivery. Physiologia 
Plantarum 164, 378–384. 

 
Atul, T., and Ray, S. (2014). Certain medicinal plants of Solanaceae and their alkaloids 

screening. International Research Journal of Medical Sciences 2, 4–6. 

 
Aversano, R., Contaldi, F., Ercolano, M. R., Grosso, V., Iorizzo, M., Tatino, F., et al. (2015). 
The Solanum commersonii Genome Sequence Provides Insights into Adaptation to Stress 
Conditions and Genome Evolution of Wild Potato Relatives. The Plant Cell 27, 954–968. 

 
Ballvora, A., Ercolano, M. R., Weiß, J., Meksem, K., Bormann, C. A., Oberhagemann, P., et al. 
(2002). The R1 gene for potato resistance to late blight (Phytophthora infestans) belongs to 
the leucine zipper/NBS/LRR class of plant resistance genes. Plant Journal 30, 361–371. 

 
Bandillo, N., Raghavan, C., Muyco, P. A., Sevilla, M. A. L., Lobina, I. T., Dilla-Ermita, C. J., et 

al. (2013). Multi-parent advanced generation inter-cross (MAGIC) populations in rice: 
progress and potential for genetics research and breeding. Rice 6, 11. 

 
Bastien, M., Boudhrioua, C., Fortin, G., and Belzile, F. (2018). Exploring the potential and 
limitations of genotyping-by-sequencing for SNP discovery and genotyping in tetraploid 
potato. Genome 61, 449–456. 

 
Belfort, M., and Bonocora, R. P. (2014). Homing endonucleases: from genetic anomalies to 

programmable genomic clippers. Methods in molecular biology (Clifton, N.J.) 1123, 1–26. 

 
Beyene, Y., Semagn, K., Mugo, S., Tarekegne, A., Babu, R., Meisel, B., et al. (2015). Genetic 
gains in grain yield through genomic selection in eight bi-parental maize populations under 
drought stress. Crop Science 55, 154–163. 

 

 

27 

Bogdanove, A. J., and Voytas, D. F. (2011). TAL Effectors: Customizable Proteins for DNA 

Targeting. Science 333, 1843 LP – 1846. 

 
Bortesi, L., and Fischer, R. (2015). The CRISPR/Cas9 system for plant genome editing and 

beyond. Biotechnol. Adv. 33, 41–52. 

 
Boudhrioua, C., Bastien, M., Légaré, G., Pomerleau, S., St-Cyr, J., Boyle, B., et al. (2017). 

Genotyping-by-Sequencing in Potato. in The Potato Genome, eds. S. Kumar Chakrabarti, C. 
Xie, and J. Kumar Tiwari (Cham: Springer International Publishing), 283–296. 

 
Boyd, L. A., Ridout, C., O’Sullivan, D. M., Leach, J. E., and Leung, H. (2013). Plant–pathogen 

interactions: disease resistance in modern agriculture. Trends in Genetics 29, 233–240. 

 
Braun, S., Gevens, A., Charkowski, A., Allen, C., and Jansky, S. (2017). Potato common scab: a 
review of the causal pathogens, management practices, varietal resistance screening methods, 
and host resistance. American Journal of Potato Research 94, 283–296. 

 
Bush, W. S., and Moore, J. H. (2012). Chapter 11: Genome-wide association studies. PLoS 

Computational Biology 8, e1002822. 

 
Butler, N. M., Atkins, P. A., Voytas, D. F., and Douches, D. S. (2015). Generation and 

inheritance of targeted mutations in potato (Solanum tuberosum L.) using the CRISPR/Cas 
system. PloS One 10, e0144591. 

 
Carroll, C. P. (1975). The inheritance and expression of sterility in hybrids of dihaploid and 

cultivated diploid potatoes. Genetica 45, 149–162. 

 
Chaparro, J. M., Holm, D. G., Broeckling, C. D., Prenni, J. E., and Heuberger, A. L. (2018). 

Metabolomics and ionomics of potato tuber reveals an influence of cultivar and market class 
on human nutrients and bioactive compounds. Frontiers in nutrition 5, 36. 

 
Charfeddine, M., Saïdi, M. N., Charfeddine, S., Hammami, A., and Gargouri Bouzid, R. (2015a). 

Genome-wide analysis and expression profiling of the erf transcription factor family in 
potato (Solanum tuberosum L.). Molecular Biotechnology 57, 348–358. 

 
Charfeddine, S., Saïdi, M. N., Charfeddine, M., and Gargouri-Bouzid, R. (2015b). Genome-wide 

identification and expression profiling of the late embryogenesis abundant genes in potato 
with emphasis on dehydrins. Molecular Biology Reports 42, 1163–1174. 

 
Chen, J. S., Ma, E., Harrington, L. B., Da Costa, M., Tian, X., Palefsky, J. M., et al. (2018). 
CRISPR-Cas12a target binding unleashes indiscriminate single-stranded DNase activity. 
Science 360, 436 LP – 439. 

 
Chen, Y., and Halterman, D. A. (2017). Phytophthora infestans effectors IPI-O1 and IPI-O4 each 

contribute to pathogen virulence. Phytopathology 107, 600–606. 

 

 

28 

Chen, Y., Liu, Z., and Halterman, D. a. (2012). Molecular determinants of resistance activation 

and suppression by Phytophthora infestans effector IPI-O. PLoS Pathogens 8, 1–10. 

 
Clasen, B. M., Stoddard, T. J., Luo, S., Demorest, Z. L., Li, J., Cedrone, F., et al. (2016). 

Improving cold storage and processing traits in potato through targeted gene knockout. Plant 
Biotechnology Journal 14, 169–176. 

 
Collard, B. C. Y., and Mackill, D. J. (2008). Marker-assisted selection: an approach for precision 
plant breeding in the twenty-first century. Philos Trans R Soc Lond B Biol Sci 363, 557–572. 

 
Collins, A., Milbourne, D., Ramsay, L., Meyer, R., Chatot-Balandras, C., Oberhagemann, P., et 

al. (1999). QTL for field resistance to late blight in potato are strongly correlated with 
maturity and vigor. Molecular Breeding 5, 387–398. 

 
Cook, D. E., Mesarich, C. H., and Thomma, B. P. H. J. (2015). Understanding plant immunity as 

a surveillance system to detect invasion. Annual Review of Phytopathology 53, 541–563. 

 
Corander, J., and Tang, J. (2007). Bayesian analysis of population structure based on linked 

molecular information. Mathematical Biosciences 205, 19–31. 

 
Couto, D., and Zipfel, C. (2016). Regulation of pattern recognition receptor signaling in plants. 

Nature Reviews Immunology 16, 537. 

 
Cox, D. B. T., Gootenberg, J. S., Abudayyeh, O. O., Franklin, B., Kellner, M. J., Joung, J., et al. 

(2017). RNA editing with CRISPR-Cas13. Science 358, 1019 LP – 1027. 

 
D’hoop, B. B., Keizer, P. L. C., Paulo, M. J., Visser, R. G. F., van Eeuwijk, F. a, and van Eck, H. 
J. (2014). Identification of agronomically important QTL in tetraploid potato cultivars using 
a marker-trait association analysis. TAG. Theoretical and applied genetics. Theoretische und 
angewandte Genetik 127, 731–48. 

 
D’Hoop, B. B., Paulo, M. J., Mank, R. A., Van Eck, H. J., and Van Eeuwijk, F. A. (2008). 

Association mapping of quality traits in potato (Solanum tuberosum L.). Euphytica 161, 47–
60. 

 
Dees, M. W., and Wanner, L. A. (2012). In search of better management of potato common scab. 

Potato Research 55, 249–268. 

 
Desta, Z. A., and Ortiz, R. (2014). Genomic selection: Genome-wide prediction in plant 

improvement. Trends in Plant Science 19, 592–601. 

 
Devaux, A., Kromann, P., and Ortiz, O. (2014). Potatoes for sustainable global food security. 

Potato Research 57, 185–199. 

 
Doudna, J. A., and Charpentier, E. (2014). The new frontier of genome engineering with 

CRISPR-Cas9. Science 346, 1258096. 

 

29 

 
Drewnowski, A., and Rehm, C. D. (2013). Vegetable cost metrics show that potatoes and beans 

provide most nutrients per penny. PloS One 8, e63277. 

 
El-Kharbotly, A., Palomino-Sánchez, C., Salamini, F., Jacobsen, E., and Gebhardt, C. (1996). R6 
and R7 alleles of potato conferring race-specific resistance to Phytophthora infestans (Mont.) 
de Bary identified genetic loci clustering with the R3 locus on chromosome XI. Theor. Appl. 
Genet 92, 880–884. 

 
Ellis, D., Chavez, O., Coombs, J., Soto, J., Gomez, R., Douches, D., et al. (2018). Genetic 
identity in genebanks: application of the SolCAP 12K SNP array in fingerprinting and 
diversity analysis in the global in trust potato collection. Genome 61, 523–537. 

 
Endelman, J. B., and Jansky, S. H. (2016). Genetic mapping with an inbred line-derived F2 

population in potato. Theoretical and Applied Genetics 129, 935–943. 

 
FAOSTAT. (2019). FAOSTAT database collections. Food and Agriculture Organization of the 

United Nations. Available at: http://www.fao.org/faostat/ [Accessed January 27, 2019]. 

 
Fasano, C., Diretto, G., Aversano, R., D’Agostino, N., Di Matteo, A., Frusciante, L., et al. 
(2016). Transcriptome and metabolome of synthetic Solanum autotetraploids reveal key 
genomic stress events following polyploidization. New Phytologist 210, 1382–1394. 

 
Felcher, K. J., Coombs, J. J., Massa, A. N., Hansey, C. N., Hamilton, J. P., Veilleux, R. E., et al. 

(2012a). Integration of two diploid potato linkage maps with the potato genome sequence. 
PloS One 7, e36347. 

 
Felcher, K. J., Coombs, J. J., Massa, A. N., Hansey, C. N., Hamilton, J. P., Veilleux, R. E., et al. 

(2012b). Integration of two diploid potato linkage maps with the potato genome sequence. 
PloS One 7, e36347. 

 
Flor, H. H. (1971). Current status of the gene-for-gene concept. Annual Review of 

Phytopathology 9, 275–296. 

 
Freyre, R., and Douches, D. S. (1994). Development of a model for marker-assisted selection of 

specific gravity in diploid potato across environments. Crop Science 34, 1361–1368. 

 
Friedman, M. (1997). Chemistry, biochemistry, and dietary role of potato polyphenols: A 

Review. Journal of Agricultural and Food Chemistry 45, 1523–1540. 

 
Fu, D., Xiao, M., Hayward, A., Fu, Y., Liu, G., Jiang, G., et al. (2014). Utilization of crop 

heterosis: a review. Euphytica 197, 161–173. 

 
Fujii, S., Kubo, K. I., and Takayama, S. (2016). Non-self- and self-recognition models in plant 

self-incompatibility. Nat. Plants 2, 16130. 

 

 

30 

Gaiero, P., Speranza, P., and de Jong, H. (2018). Introgressive hybridization in potato revealed 
by novel cytogenetic and genomic technologies. American Journal of Potato Research 95, 
607–621. 

 
Gálvez, J. H., Tai, H. H., Lagüe, M., Zebarth, B. J., and Strömvik, M. V (2016). The nitrogen 

responsive transcriptome in potato (Solanum tuberosum L.) reveals significant gene 
regulatory motifs. Scientific Reports 6, 26090. 

 
Gebhardt, C., Ballvora, A., Walkemeier, B., Oberhagemann, P., and Schüler, K. (2004). 
Assessing genetic potential in germplasm collections of crop plants by marker-trait 
association: A case study for potatoes with quantitative variation of resistance to late blight 
and maturity type. Molecular Breeding 13, 93–102. 

 
George, T. S., Taylor, M. A., Dodd, I. C., and White, P. J. (2017). Climate change and 

consequences for potato production: a review of tolerance to emerging abiotic stress. Potato 
Research 60, 239–268. 

 
Goldraij, A., Kondo, K., Lee, C. B., Hancock, C. N., Sivaguru, M., Vazquez-Santana, S., et al. 

(2006). Compartmentalization of S-RNase and HT-B degradation in self-incompatible 
Nicotiana. Nature 439, 805. 

 
Gong, L., Zhang, H., Gan, X., Zhang, L., Chen, Y., Nie, F., et al. (2015). Transcriptome profiling 

of the potato (Solanum tuberosum L.) plant under drought stress and water-stimulus 
Conditions. PloS One 10, e0128041. 

 
Grada, C. (2004). Ireland’s great famine: An overview. Dublin. No.04/24 
 
Grun, P. (1990). The evolution of cultivated potatoes. Economic Botany 44, 39–55. 
 
Grünwald, N. J., and Flier, W. G. (2005). The biology of Phytophthora infestans at Its Center of 

Origin. Annual Review of Phytopathology 43, 171–190. 

 
Habyarimana, E., Parisi, B., and Mandolino, G. (2017). Genomic prediction for yields, 

processing and nutritional quality traits in cultivated potato (Solanum tuberosum L.). Plant 
Breeding 136, 245–252. 

 
Hafez, M., and Hausner, G. (2012). Homing endonucleases: DNA scissors on a mission. Genome 

55, 553–569. 

 
Halterman, D. A., Chen, Y., Sopee, J., Berduo-Sandoval, J., and Sánchez-Pérez, A. (2010). 

Competition between Phytophthora infestans effectors leads to increased aggressiveness on 
plants containing broad-spectrum late blight resistance. PloS One 5, e10536. 

 
Halterman, D. a, Kramer, L. C., Wielgus, S., and Jiang, J. (2008). Performance of transgenic 

potato containing the late blight resistance gene RB. Plant Disease 92, 339–343. 

 

 

31 

Hämäläinen, J. H., Watanabe, K. N., Valkonen, J. P. T., Arihara, a., Plaisted, R. L., Pehu, E., et 
al. (1997). Mapping and marker-assisted selection for a gene for extreme resistance to potato 
virus Y. Theor Appl Genet, 192–197. 

 
Hardigan, M. A., Crisovan, E., Hamiltion, J. P., Kim, J., Laimbeer, P., Leisner, C. P., et al. 

(2016). Genome reduction uncovers a large dispensable genome and adaptive role for copy 
number variation in asexually propagated Solanum tuberosum. Plant Cell 28, 388–405. 

 
Hardigan, M. A., Laimbeer, F. P. E., Newton, L., Crisovan, E., Hamilton, J. P., Vaillancourt, B., 

et al. (2017). Genome diversity of tuber-bearing Solanum uncovers complex evolutionary 
history and targets of domestication in the cultivated potato. Proceedings of the National 
Academy of Sciences 114, E9999 LP-E10008. 

 
Haverkort, A. J., Struik, P. C., Visser, R. G. F., and Jacobsen, E. (2009). Applied biotechnology 
to combat bate blight in potato caused by Phytophthora Infestans. Potato Research 52, 249–
264. 

 
Hawkes, J. G. (1958). Significance of wild species and primitive forms for potato breeding. 

Euphytica 7, 257–270. 

 
Hawkes, J. G. (1992). “History of the potato,” in The potato crop: the scientific basis for 

improvement, ed. P. Harris (Dordrecht: Springer Netherlands), 1–13. 

 
Hawkes, J. G., and Francisco-Ortega, J. (1993). The early history of the potato in Europe. 

Euphytica 70, 1–7. 

 
Hayes, B. J., Bowman, P. J., Chamberlain, A. J., and Goddard, M. E. (2009). Genomic selection 

in dairy cattle: progress and challenges. Journal of Dairy Science 92, 433–443. 

 
Heffner, E. L., Sorrells, M. E., and Jannink, J.-L. (2009). Genomic selection for crop 

improvement. Crop Science 49, 1–12. 

 
Hoffman, G. E. (2013). Correcting for population structure and kinship using the linear mixed 

model: Theory and extensions. PLoS One 8, e75707. 

 
Hosaka, K. (2003). T-type chloroplast DNA in Solanum tuberosum L. ssp. tuberosum was 
conferred from some populations of S. tarijense Hawkes. American Journal of Potato 
Research 80, 21–32. 

 
Hosaka, K., and Hanneman, R. E. (1988). The origin of the cultivated tetraploid potato based on 

chloroplast DNA. Theoretical and Applied Genetics 76, 172–176. 

 
Hosaka, K., and Hanneman, R. E. (1998). Genetics of self-compatibility in a self-incompatible 

wild diploid potato species Solanum chacoense. 1. Detection of an S locus inhibitor (Sli) 
gene. Euphytica 99, 191–197. 

 

 

32 

Hu, J. H., Miller, S. M., Geurts, M. H., Tang, W., Chen, L., Sun, N., et al. (2018). Evolved Cas9 

variants with broad PAM compatibility and high DNA specificity. Nature 556, 57. 

 
Hua, Z.-H., Fields, A., and Kao, T. (2008). Biochemical models for S-RNase-based self-

incompatibility. Molecular Plant 1, 575–585. 

 
Huamán, Z., and Spooner, D. M. (2002). Reclassification of landrace populations of cultivated 

potatoes (Solanum sect. Petota). American Journal of Botany 89, 947–965. 

 
Huang, B. E., Verbyla, K. L., Verbyla, A. P., Raghavan, C., Singh, V. K., Gaur, P., et al. (2015). 

MAGIC populations in crops: current status and future prospects. Theoretical and Applied 
Genetics 128, 999–1017. 

 
Iovene, M., Zhang, T., Lou, Q., Buell, C. R., and Jiang, J. (2013). Copy number variation in 

potato – an asexually propagated autotetraploid species. The Plant Journal 75, 80–89. 

 
Jaganathan, D., Ramasamy, K., Sellamuthu, G., Jayabalan, S., and Venkataraman, G. (2018). 

CRISPR for crop improvement: An update review. Front. Plant. Sci. 9, 985. 

 
Jansky, S. H., Charkowski, A. O., Douches, D. S., Gusmini, G., Richael, C., Bethke, P. C., et al. 
(2016). Reinventing potato as a diploid inbred line–based crop. Crop Science 56, 1412–1422. 

 
Jansky, S. H., Chung, Y. S., and Kittipadukal, P. (2014). M6: A diploid potato inbred line for use 

in breeding and genetics research. J. Plant. Regist. 8, 195–199. 

 
Jansky, S. H., and Peloquin, S. J. (2006). Advantages of wild diploid Solanum species over 

cultivated diploid relatives in potato breeding programs. Genetic Resources and Crop 
Evolution 53, 669–674. 

 
Jiménez-Durán, K., McClure, B., García-Campusano, F., Rodríguez-Sotres, R., Cisneros, J., 

Busot, G., et al. (2013). NaStEP: A proteinase inhibitor essential to self-incompatibility and a 
positive regulator of HT-B stability in Nicotiana alata pollen tubes. Plant Physiology 161, 97 
LP – 107. 

 
Jones, J. D. G., and Dangl, J. L. (2006). The plant immune system. Nature 444, 323–329.  

Joung, J. K., and Sander, J. D. (2013). TALENs: a widely applicable technology for targeted 
genome editing. Nature reviews. Molecular cell biology 14, 49–55. 

 
Jupe, F., Witek, K., Verweij, W., Śliwka, J., Pritchard, L., Etherington, G. J., et al. (2013). 

Resistance gene enrichment sequencing (RenSeq) enables reannotation of the NB-LRR gene 
family from sequenced plant genomes and rapid mapping of resistance loci in segregating 
populations. Plant Journal 76, 530–544. 

 
Kloosterman, B., Abelenda, J. A., Gomez, M. del M. C., Oortwijn, M., de Boer, J. M., 

Kowitwanich, K., et al. (2013). Naturally occurring allele diversity allows potato cultivation 
in northern latitudes. Nature 495, 246. 

 

33 

 
Kolasa, K. (1993). The potato and human nutrition. American Potato Journal 70, 375–384. 
 
Kromann, P., Miethbauer, T., Ortiz, O., and Forbes, G. A. (2014). “Review of potato biotic 

constraints and experiences with integrated pest management interventions,” in Integrated 
Pest Management. Pesticide problems, Vol.3, eds. D. Pimentel and R. Peshin (Dordrecht: 
Springer Netherlands), 245–268. 

 
Kubo, K., Entani, T., Takara, A., Wang, N., Fields, A. M., Hua, Z., et al. (2010). Collaborative 
non-self recognition system in S-RNase based self-incompatibility. Science 330, 796–799. 

 
Kubo, K., Paape, T., Hatakeyama, M., Entani, T., Takara, A., Kajihara, K., et al. (2015). Gene 
duplication and genetic exchange drive the evolution of S-RNase-based self-incompatibility 
in Petunia. Nat. Plants 1, 14005. 

 
Kuhl, J. C., Hanneman Jr., R. E., and Havey, M. J. (2001). Characterization of and mapping of 
Rpi1, a late-blight resistance locus from diploid (1EBN) mexican Solanum pinnatisectum. 
Molecular Genetics and Genomics 265, 977–985. 

 
Kusano, H., Ohnuma, M., Mutsuro-Aoki, H., Asahi, T., Ichinosawa, D., Onodera, H., et al. 
(2018). Establishment of a modified CRISPR/Cas9 system with increased mutagenesis 
frequency using the translational enhancer dMac3 and multiple guide RNAs in potato. 
Scientific Reports 8, 13753. 

 
Lee, C. B., Kim, S., and McClure, B. (2009). A pollen protein, NaPCCP, that binds pistil 

arabinogalactan proteins also binds phosphatidylinositol 3-phosphate and associates with the 
pollen tube endomembrane system. Plant Physiology 149, 791 LP – 802. 

 
Leesutthiphonchai, W., Vu, A. L., Ah-Fong, A. M. V, and Judelson, H. S. (2018). How does 
Phytophthora infestans evade control efforts? Modern insight into the late blight disease. 
Phytopathology 108, 916–924. 

 
Leisner, C. P., Hamilton, J. P., Crisovan, E., Manrique‐Carpintero, N. C., Marand, A. P., 
Newton, L., et al. (2018). Genome sequence of M6, a diploid inbred clone of the high‐
glycoalkaloid‐producing tuber‐bearing potato species Solanum chacoense, reveals residual 
heterozygosity. 

 
Li, Y., Colleoni, C., Zhang, J., Liang, Q., Hu, Y., Ruess, H., et al. (2018). Genomic analyses 

yield markers for identifying agronomically important genes in potato. Molecular Plant 11, 
473–484. 

 
Lian, Q., Tang, D., Bai, Z., Qi, J., Lu, F., Huang, S., et al. (2019). Acquisition of deleterious 
mutations during potato polyploidization. Journal of Integrative Plant Biology 61, 7–11. 

 

 

34 

Lipka, A. E., Kandianis, C. B., Hudson, M. E., Yu, J., Drnevich, J., Bradbury, P. J., et al. (2015). 

From association to prediction: statistical methods for the dissection and selection of 
complex traits in plants. Current Opinion in Plant Biology 24, 110–118. 

 
Lister, A., Heavens, D., Baker, D., Cugliandolo, F., Kettleborough, G., Macaulay, I., et al. 
(2019). A critical comparison of technologies for a plant genome sequencing project. 
GigaScience, giy163. doi:10.1093/gigascience/giy163. 

 
Lozano, R., Ponce, O., Ramirez, M., Mostajo, N., and Orjeda, G. (2012). Genome-wide 

identification and mapping of NBS-Encoding Resistance Genes in Solanum tuberosum 
Group Phureja. PLoS One 7, e34775. 

 
Malosetti, M., Van Der Linden, C. G., Vosman, B., and Van Eeuwijk, F. A. (2007). A mixed-
model approach to association mapping using pedigree information with an illustration of 
resistance to Phytophthora infestans in potato. Genetics 175, 879–889. 

 
Manrique-Carpintero, N. C., Coombs, J. J., Cui, Y., Veilleux, R. E., Buell, C. R., and Douches, 
D. (2015). Genetic map and QTL analysis of agronomic traits in a diploid potato population 
using single nucleotide polymorphism markers. Crop Science 55, 2566–2579. 

 
Mariot, R. F., de Oliveira, L. A., Voorhuijzen, M. M., Staats, M., Hutten, R. C. B., van Dijk, J. 

P., et al. (2016). Characterization and transcriptional profile of genes involved in 
glycoalkaloid biosynthesis in new varieties of Solanum tuberosum L. Journal of Agricultural 
and Food Chemistry 64, 988–996. 

 
Massa, A. N., Childs, K. L., and Buell, C. R. (2013). Abiotic and biotic stress responses in 

Solanum tuberosum Group Phureja DM1-3 516 R44 as measured through whole 
transcriptome Sequencing. The Plant Genome 6. doi:10.3835/plantgenome2013.05.0014. 

 
Massa, A. N., Childs, K. L., Lin, H., Bryan, G. J., Giuliano, G., and Buell, C. R. (2011). The 
transcriptome of the reference Potato Genome Solanum tuberosum Group Phureja clone 
DM1-3 516R44. PloS One 6, e26801. 

 
Massa, A. N., Manrique-Carpintero, N. C., Coombs, J. J., Zarka, D. G., Boone, A. E., Kirk, W. 

W., et al. (2015). Genetic linkage mapping of economically important traits in cultivated 
tetraploid potato (Solanum tuberosum L.). G3: Genes, Genomes, Genetics 5, 2357–2364. 

 
McClure, B. A., Haring, V., Ebert, P. R., Anderson, M. A., Simpson, R. J., Sakiyama, F., et al. 

(1989). Style self-incompatibility gene products of Nicotlana alata are ribonucleases. Nature 
342, 955–957. 

 
McClure, B., Mou, B., Canevascini, S., and Bernatzky, R. (1999). A small asparagine-rich 

protein required for S-allele-specific pollen rejection in Nicotiana. Proc. Natl. Acad. Sci. U S 
A. 96, 13548–13553. 

 

35 

Mendoza, H. A., Mihovilovich, E. J., and Saguma, F. (1996). Identification of triplex (YYYy) 

Potato Virus Y (PVY) immune progenitors derived from Solanum tuberosum ssp. andigena. 
Am. Potato J. 73, 13–19. 

 
Meuwissen, T. H. E., Hayes, B. J., and Goddard, M. E. (2001). Prediction of total genetic value 

using genome-wide dense marker maps. Genetics 157, 1819–1829. 

 
Miller, J. C., Holmes, M. C., Wang, J., Guschin, D. Y., Lee, Y.-L., Rupniewski, I., et al. (2007). 

An improved zinc-finger nuclease architecture for highly specific genome editing. Nature 
Biotechnology 25, 778. 

 
Nakayasu, M., Akiyama, R., Lee, H. J., Osakabe, K., Osakabe, Y., Watanabe, B., et al. (2018). 

Generation of α-solanine-free hairy roots of potato by CRISPR/Cas9 mediated genome 
editing of the St16DOX gene. Plant Physiology and Biochemistry 131, 70–77. 

 
National Potato Council (2016). Potato statistical year book. Washington D.C. 
 
Nicolia, A., Proux-Wéra, E., Åhman, I., Onkokesung, N., Andersson, M., Andreasson, E., et al. 
(2015). Targeted gene mutation in tetraploid potato through transient TALEN expression in 
protoplasts. Journal of Biotechnology 204, 17–24. 
 
Nishimasu, H., Shi, X., Ishiguro, S., Gao, L., Hirano, S., Okazaki, S., et al. (2018). Engineered 

CRISPR-Cas9 nuclease with expanded targeting space. Science 361, 1259 LP – 1262. 

 
Nowicki, M., Foolad, M. R., Nowakowska, M., and Kozik, E. U. (2011). Potato and tomato late 
blight caused by Phytophthora infestans: An overview of pathology and resistance breeding. 
Plant Disease 96, 4–17. 

 
O’Brien, M., Kapfer, C., Major, G., Laurin, M., Bertrand, C., Kondo, K., et al. (2002). Molecular 

analysis of the stylar-expressed Solanum chacoense small asparagine-rich protein family 
related to the HT modifier of gametophytic self-incompatibility in Nicotiana. Plant J 32, 
985–996.  

 
O’Brien, M., Major, G., Chantha, S.-C., and Matton, D. P. (2004). Isolation of S-RNase binding 

proteins from Solanum chacoense: identification of an SBP1 (RING finger protein) 
orthologue. Sexual Plant Reproduction 17, 81–87. 

 
Oerke, E. C. (2006). Crop losses to pests. The Journal of Agricultural Science 144, 31–43. 
 
Olmstead, R. G., Bohs, L., Migid, H. A., Santiago-valentin, E., Garcia, V. F., and Collier, S. M. 
(2008). A molecular phylogeny of the Solanaceae. Molecular phylogenetics 57, 1159–1181. 

 
Ovchinnikova, A., Krylova, E., Gavrilenko, T., Smekalova, T., Zhuk, M., Knapp, S., et al. 

(2011). Taxonomy of cultivated potatoes (Solanum section Petota: Solanaceae). Botanical 
Journal of the Linnean Society 165, 107–155. 

 

 

36 

Paajanen, P., Kettleborough, G., Lopez-Girona, E., Giolai, M., Heavens, D., Baker, D., et al. 

(2017). A critical comparison of technologies for a plant genome sequencing project. 
bioRxiv, 201830. doi:10.1101/201830. 

 
Park, T.-H., Gros, J., Sikkema, A., Vleeshouwers, V. G., Muskens, M., Allefs, S., et al. (2005a). 

The late blight resistance locus Rpi-bib3 from Solanum bulbocastanum belongs to a major 
late blight R gene cluster on chromosome 4 of potato. Molecular plant-microbe interactions: 
MPMI 18, 722–729. 

 
Park, T.-H., Vleeshouwers, V. G. A. A., Hutten, R. C. B., van Eck, H. J., van der Vossen, E., 
Jacobsen, E., et al. (2005b). High-resolution mapping and analysis of the resistance locus 
Rpi-abpt against Phytophthora infestans in Potato. Molecular Breeding 16, 33–43. 
doi:10.1007/s11032-005-1925-z. 

 
Pawluk, A., Davidson, A. R., and Maxwell, K. L. (2017). Anti-CRISPR: discovery, mechanism 

and function. Nature Reviews Microbiology 16, 12. 

 
Pel, M. a, Foster, S. J., Park, T.-H., Rietman, H., van Arkel, G., Jones, J. D. G., et al. (2009). 

Mapping and cloning of late blight resistance genes from Solanum venturii using an 
interspecific candidate gene approach. Molecular plant-microbe interactions: MPMI 22, 601–
615. 

 
Pellagatti, A., Dolatshad, H., Valletta, S., and Boultwood, J. (2015). Application of 

CRISPR/Cas9 genome editing to the study and treatment of disease. Arch. Toxicol. 89, 
1023–1034. 

 
Pham, G. M., Newton, L., Wiegert-Rininger, K., Vaillancourt, B., Douches, D. S., and Buell, C. 
R. (2017). Extensive genome heterogeneity leads to preferential allele expression and copy 
number-dependent expression in cultivated potato. The Plant Journal 92, 624–637. 

 
Piñeros-Niño, C., Narváez-Cuenca, C.-E., Kushalappa, A. C., and Mosquera, T. (2017). 

Hydroxycinnamic acids in cooked potato tubers from Solanum tuberosum group Phureja. 
Food Science & Nutrition 5, 380–389. 

 
Poland, J., and Rutkoski, J. (2016). Advances and Challenges in Genomic Selection for Disease 

Resistance. Annual Review of Phytopathology 54, 79–98. 

 
Pomerantz, A., Cohen, Y., Shufan, E., Ben-Naim, Y., Mordechai, S., Salman, A., et al. (2014). 

Characterization of Phytophthora infestans resistance to mefenoxam using FTIR 
spectroscopy. Journal of Photochemistry and Photobiology B: Biology 141, 308–314. 

 
Rees, H. A., and Liu, D. R. (2018). Base editing: precision chemistry on the genome and 

transcriptome of living cells. Nature Reviews Genetics 19, 770–788. 

 
Rockman, M. V, and Kruglyak, L. (2008). Breeding designs for recombinant inbred advanced 

intercross lines. Genetics 179, 1069–1078. 

 

37 

 
Rodríguez-Pérez, C., Gómez-Caravaca, A. M., Guerra-Hernández, E., Cerretani, L., García-
Villanova, B., and Verardo, V. (2018). Comprehensive metabolite profiling of Solanum 
tuberosum L. (potato) leaves by HPLC-ESI-QTOF-MS. Food Research International 112, 
390–399. 

 
Rosyara, U. R., De Jong, W. S., Douches, D. S., and Endelman, J. B. (2016). Software for 

genome-wide association studies in autopolyploids and its application to potato. The Plant 
Genome 9, 1–10. 

 
Rumold, C. U., and Aldenderfer, M. S. (2016). Late Archaic-Early Formative period 

microbotanical evidence for potato at Jiskairumoko in the Titicaca Basin of southern Peru. 
Proceedings of the National Academy of Sciences 113, 13672–13677. 

 
Sassa, H., Kakui, H., Miyamoto, M., Suzuki, Y., Hanada, T., Ushijima, K., et al. (2007). S locus 

F-box brothers: multiple and pollen-specific F-box genes with S haplotype-specific 
polymorphisms in apple and Japanese pear. Genetics 175, 1869–1881. 

 
Schönhals, E. M., Ortega, F., Barandalla, L., Aragones, A., Ruiz de Galarreta, J. I., Liao, J.-C., et 
al. (2016). Identification and reproducibility of diagnostic DNA markers for tuber starch and 
yield optimization in a novel association mapping population of potato (Solanum tuberosum 
L.). Theoretical and Applied Genetics 129, 767–785. 

 
Sharma, S. K., Bolser, D., de Boer, J., Sønderkær, M., Amoros, W., Carboni, M. F., et al. (2013). 

Construction of Reference Chromosome-Scale Pseudomolecules for Potato: Integrating the 
Potato Genome with Genetic and Physical Maps. G3: Genes|Genomes|Genetics 3, 2031 LP – 
2047. 

 
Sijacic, P., Wang, X., Skirpan, A. L., Wang, Y., Dowd, P. E., McCubbin, A. G., et al. (2004). 
Identification of the pollen determinant of S-RNase-mediated self-incompatibility. Nature 
429, 302–305. 

 
Simmonds, N. W. (1997). A review of potato propagation by means of seed, as distinct from 

clonal propagation by tubers. Potato Research 40, 191–214. 

 
Slater, A. T., Cogan, N. O. I., Forster, J. W., Hayes, B. J., and Daetwyler, H. D. (2016). 

Improving genetic gain with genomic selection in autotetraploid potato. The Plant Genome 9, 
1–15. 

 
Slater, A. T., Wilson, G. M., Cogan, N. O. I., Forster, J. W., and Hayes, B. J. (2014). Improving 

the analysis of low heritability complex traits for enhanced genetic gain in potato. TAG. 
Theoretical and applied genetics. Theoretische und angewandte Genetik 127, 809–20. 

 
Smart, C. D., and Fry, W. E. (2001). Invasions by the late blight pathogen: Renewed sex and 

enhanced fitness. Biological Invasions 3, 235–243. 

 

 

38 

Song, J., Bradeen, J. M., Naess, S. K., Raasch, J. a, Wielgus, S. M., Haberlach, G. T., et al. 

(2003). Gene RB cloned from Solanum bulbocastanum confers broad spectrum resistance to 
potato late blight. Proceedings of the National Academy of Sciences of the United States of 
America 100, 9128–9133. 

 
Song, S., Hao, L., Zhao, P., Xu, Y., Zhong, N., Zhang, H., et al. (2019). Genome-wide 

Identification, Expression Profiling and Evolutionary Analysis of Auxin Response Factor 
Gene Family in Potato (Solanum tuberosum Group Phureja). Scientific Reports 9, 1755. 

 
Spindel, J. E., Begum, H., Akdemir, D., Collard, B., Redoña, E., Jannink, J.-L., et al. (2016). 

Genome-wide prediction models that incorporate de novo GWAS are a powerful new tool for 
tropical rice improvement. Heredity 116, 395. 

 
Spooner, D. M., Sytsma, K. J., and Conti, E. (1991). Chloroplast dna evidence for genome 
differentiation in wild potatoes (Solanum sect. petota: Solanaceae). American Journal of 
Botany 78, 1354–1366. 

 
Sprenger, H., Kurowsky, C., Horn, R., Erban, A., Seddig, S., Rudack, K., et al. (2016). The 
drought response of potato reference cultivars with contrasting tolerance. Plant, Cell & 
Environment 39, 2370–2389. 

 
Sternberg, S. H., Redding, S., Jinek, M., Greene, E. C., and Doudna, J. A. (2014). DNA 

interrogation by the CRISPR RNA-guided endonuclease Cas9. Nature 507, 62–7. 

 
Sukarta, O. C. A., Slootweg, E. J., and Goverse, A. (2016). Structure-informed insights for NLR 

functioning in plant immunity. Seminars in Cell & Developmental Biology 56, 134–149. 
 

Sverrisdóttir, E., Byrne, S., Sundmark, E. H. R., Johnsen, H. Ø., Kirk, H. G., Asp, T., et al. 

(2017). Genomic prediction of starch content and chipping quality in tetraploid potato using 
genotyping-by-sequencing. Theoretical and Applied Genetics 130, 2091–2108. 
 

Takayama, S., and Isogai, A. (2005). Self-Incompatibility in Plants. Annu. Rev. Plant. Biol. 56, 

467–489. 

 
The Potato Genome Sequencing Consortium (2011). Genome sequence and analysis of the tuber 

crop potato. Nature 475, 189–195. 

 
Tomita, S., Ikeda, S., Tsuda, S., Someya, N., Asano, K., Kikuchi, J., et al. (2017). A survey of 

metabolic changes in potato leaves by NMR-based metabolic profiling in relation to 
resistance to late blight disease under field conditions. Magnetic Resonance in Chemistry 55, 
120–127. 

 
Urnov, F. D., Rebar, E. J., Holmes, M. C., Zhang, H. S., and Gregory, P. D. (2010). Genome 

editing with engineered zinc finger nucleases. Nature Reviews Genetics 11, 636. 

 

 

39 

Vinson, J. a., Demkosky, C. a., Navarre, D. a., and Smyda, M. a. (2012). High-antioxidant 

potatoes: Acute in vivo antioxidant source and hypotensive agent in humans after 
supplementation to hypertensive subjects. Journal of Agricultural and Food Chemistry 60, 
6749–6754. 

 
Vos, P. G., Uitdewilligen, J. G. A. M. L., Voorrips, R. E., Visser, R. G. F., and van Eck, H. J. 
(2015). Development and analysis of a 20K SNP array for potato (Solanum tuberosum): an 
insight into the breeding history. TAG Theoretical and applied genetics. 128, 2387–2401. 

 
Vossen, E. A. G., Gros, J., Sikkema, A., Muskens, M., Wouters, D., Wolters, P., et al. (2005). 

The Rpi-blb2 gene from Solanum bulbocastanum is an Mi-1 gene homolog conferring broad-
spectrum late blight resistance in potato. The Plant Journal 44, 208–222. 

 
Wang, S., Zhang, S., Wang, W., Xiong, X., Meng, F., and Cui, X. (2015). Efficient targeted 

mutagenesis in potato by the CRISPR/Cas9 system. Plant Cell Reports. 

 
Wanner, L. A. (2006). A survey of genetic variation in Streptomyces isolates causing potato 

common scab in the United States. Phytopathology 96, 1363–1371. 

 
Wanner, L. A., and Kirk, W. W. (2015). Streptomyces – from basic microbiology to role as a 

plant pathogen. American Journal of Potato Research 92, 236–242. 

 
Weese, T. L., and Bohs, L. (2007). A Three-Gene Phylogeny of the Genus Solanum 

(Solanaceae). Systematic Botany 32, 445–463. 

 
Witek, K., Jupe, F., Witek, A. I., Baker, D., Clark, M. D., and Jones, J. D. G. (2016). Accelerated 
cloning of a potato late blight–resistance gene using RenSeq and SMRT sequencing. Nature 
Biotechnology 34, 656–660. 

 
Wu, F., and Tanksley, S. D. (2010). Chromosomal evolution in the plant family Solanaceae. 

BMC genomics 11, 182. 

 
Ye, M., Peng, Z., Tang, D., Yang, Z., Li, D., Xu, Y., et al. (2018). Generation of self-compatible 

diploid potato by knockout of S-RNase. Nat. Plants 4, 651–654. 

 
Yogendra, K. N., Kushalappa, A. C., Sarmiento, F., Rodriguez, E., and Mosquera, T. (2015). 

Metabolomics deciphers quantitative resistance mechanisms in diploid potato clones against 
late blight. Functional Plant Biology 42, 284–298. 

 
Yu, J., and Buckler, E. S. (2006). Genetic association mapping and genome organization of 

maize. Current Opinion in Biotechnology 17, 155–160. 

 
Zhang, X., Pérez-Rodríguez, P., Semagn, K., Beyene, Y., Babu, R., López-Cruz, M. a, et al. 
(2014). Genomic prediction in biparental tropical maize populations in water-stressed and 
well-watered environments using low-density and GBS SNPs. Heredity, 1–9. 

 

 

40 

Zhang, Z., Ersoz, E., Lai, C.-Q., Todhunter, R. J., Tiwari, H. K., Gore, M. A., et al. (2010). 

Mixed linear model approach adapted for genome-wide association studies. Nature Genetics 
42, 355. 

 
Zhou, X., and Stephens, M. (2012). Genome-wide efficient mixed-model analysis for association 

 

studies. Nature Genetics 44, 821. 

 
Zong, Y., Song, Q., Li, C., Jin, S., Zhang, D., Wang, Y., et al. (2018). Efficient C-to-T base 

editing in plants using a fusion of nCas9 and human APOBEC3A. Nature Biotechnology 36, 
950. 

 

 

41 

GENOMIC SELECTION FOR LATE BLIGHT AND COMMON SCAB RESISTANCE 

Chapter 2  

IN TETRAPLOID POTATO (Solanum tuberosum) 

[Published in G3: Genes, Genomes, Genetics. 8(7):2471-2481] 

 

Felix Enciso-Rodriguez1, David Douches1, Marco Lopez-Cruz1, Joseph Coombs1 and Gustavo de 

Los Campos.2* 

1 Department of Plant, Soil, and Microbial Sciences, Michigan State University. East Lansing, 

Michigan,  48824.  2  Department  of  Epidemiology  &  Biostatistics,  Department  of  Statistics  & 

Probability  and  Institute  for  Quantitative  Health  Science  and  Engineering,  Michigan  State 

University. East Lansing, Michigan, 48824. 

*Corresponding author (deloscampos@epi.msu.edu) 

 

 

 

42 

I. Abstract 

Potato (Solanum tuberosum) is a staple food crop and is considered one of the main sources of 

carbohydrates worldwide. Late blight (Phytophthora infestans) and common scab (Streptomyces 

scabies) are two of the primary production constraints faced by potato farming. Previous studies 

have identified a few resistance genes for both late blight and common scab; however, these 

genes explain only a limited fraction of the heritability of these diseases. Genomic selection has 

been demonstrated to be an effective methodology for breeding value prediction in many major 

crops (e.g., maize and wheat). However, the technology has received little attention in potato 

breeding. We present the first genomic selection study involving late blight and common scab in 

tetraploid potato. Our data involves 4,110 Single Nucleotide Polymorphisms (SNPs) genetic 

markers and field evaluations for 1,763 late blight phenotypes collected in seven years and 3,885 

common scab phenotypes in nine years. We report moderately high genomic heritability 

estimates (0.46±0.04 and 0.45±0.017, for late blight and common scab, respectively). The extent 

of genotype-by-year interaction was high for late blight and low for common scab. Our 

assessment of prediction accuracy demonstrates the applicability of genomic prediction for 

tetraploid potato breeding. For both traits, we found that more than 90% of the genetic variance 

could be captured with an additive model. For common scab, the highest prediction accuracy was 

achieved using an additive model. For late blight, small but statistically significant gains in 

prediction accuracy were achieved using a model that accounted for both additive and 

dominance effects. Using whole-genome regression models we identified SNPs located in 

previously reported hotspots regions for late blight, on genes associated with systemic disease 

resistance responses, and a new locus located in a WRKY transcription factor for common scab. 

 

 

43 

II. Introduction 

The potato (Solanum tuberosum L.) is considered the sixth most important agricultural 

commodity worldwide after sugar cane, maize, rice, wheat and milk. In 2014, the global 

production of potatoes exceeded 385 million tons, positioning China as the largest producer with 

more than 66 million tons, followed by Russia, India and the United States (FAOSTAT 2016). 

As a staple food, this crop represents one of the main sources of carbohydrates, fiber, minerals 

and vitamins, providing essential nutrients and energy needed for healthy body development and 

function (Kolasa, 1993; Drewnowski and Rehm, 2013). 

Despite its great economic and food security importance, potatoes face high production 

losses caused mainly by biotic factors. Among them, pathogens such as late blight (Phytophthora 

infestans (Mont.) de Bary), represent the most devastating disease for potato worldwide. Late 

blight infects vegetative tissues, typically killing the entire plant, within 7 to 10 days. This 

pathogen accounts for annual losses of 16% of the total global potato production (Haverkort et 

al., 2009). Under increasingly variable weather conditions, late blight incidence is expected to 

escalate worldwide, affecting mainly highlands in developing countries (Sparks et al. 2014). 

Soil-borne pathogens such as common scab (Streptomyces scabies Thaxter), reduces the 

potato quality and marketability by causing superficial lesions on the tuber surface (Dees and 

Wanner 2012). Susceptibility to common scab is dependent upon genotype, time and 

environmental conditions (Wanner 2006; Wanner and Kirk 2015), having a negative impact 

mainly in underground tissues in development, such as stolons and tubers. This pathogen has 

spread worldwide and due to its saprophyte nature, common scab can survive in winter, thus 

becoming a permanent source of inoculum for the next planting seasons, causing losses up to 

$100/Ha (Wanner and Kirk 2015). 

 

44 

Although pathogen infection can be controlled by using protectants or systematic fungicides, 

they can be ineffective if the environmental conditions favor pathogen dispersion (Nowicki et al. 

2011) or the emergence of fungicide-resistant genotypes (Pomerantz et al. 2014). The most 

effective way to control the incidence of late blight and common scab in potatoes is through the 

generation of resistant varieties (Ahn and Park 2013). However, breeding for resistant varieties 

via phenotypic selection can take up to 15 years, which is time-consuming and ineffective 

against fast-evolving pathogens (The Potato Genome Sequencing Consortium 2011; Lozano et 

al. 2012). 

Marker-assisted (Barone 2004) and genomic selection (GS) strategies (Meuwissen et al. 

2001) can accelerate the process of breeding disease resistance. Several studies on late blight and 

common scab resistance have reported variants conferring resistance to these pathogens; 

however, most of the genomic research has focused on late blight (Gebhardt et al. 2004; 

Malosetti et al. 2007; Muktar et al. 2015; Mosquera et al. 2016; Braun, Endelman, et al. 2017) 

and are largely based on phenotype-single marker association analyses. To the best of our 

knowledge, no study so far has considered the use of GS for breeding resistance to late blight and 

common scab in potato. Therefore, in this article, we use Whole-Genome Regression methods 

commonly used in GS to: (i) study important features of the genetic architecture of resistance to 

late blight and common scab (including trait heritability, extent of genetic-by-environment 

interactions (G´E) and the importance of non-additive effects), (ii) identify large-effect variants 

contributing to resistance to late blight and scab, and (iii) assess the prediction accuracy of GS 

for resistance to those two pathogens. 

Our data involves (up to) nine years of field evaluations for late blight and common scab at 

two Michigan State University’s (MSU) research centers. We considered models that accounted 

 

45 

for additive effects and various forms of dominance and evaluated two different statistical 

methods. Our results suggest that sizable fraction of the inter-individual differences in disease 

resistance (~ 46% for late blight and 45% for common scab) can be captured by the SNP set used 

in the study. The extent of G´E was low for common scab and high for late blight. We found 

that additive models can capture more than 90% of the genetic variance. We report large-effect 

SNPs contributing to late blight resistance in chromosomes V and IX, that have been previously 

reported to harbor resistance genes to this pathogen. We also report the first SNP associated with 

common scab resistance, located on chromosome IX, and positioned in a transcription factor 

known for its role in systemic defense and resistance responses. Our results demonstrate that 

genomic selection can yield moderately accurate prediction of disease resistance for genotypes 

that have been not evaluated in field trials. Thus, GS could be used for rapid cycling selection for 

resistance to both late blight and common scab in tetraploid potato. 

 

III. Materials and Methods 

i. 

Data 

Data were collected from early generation and advanced tetraploid potato genotypes derived 

from bi-parental crosses at the MSU potato breeding program. Additional advanced breeding 

genotypes from other United States breeding programs and reference varieties were also 

included. The available genotypes (n=381) represent different market classes for fresh market, 

chip-processing, and russet-type fresh market and processing varieties. These genotypes were 

evaluated in field trials that included annual selections from MSU’s potato breeding program, 

where each year poorly performing genotypes were replaced with new genotypes, while 

maintaining control genotypes during consecutive years. 

 

46 

Late blight field resistance trials (273 genotypes and a total of 1,763 disease records) were 

conducted in inoculated foliar field trials during seven years (2010-2015 and 2017) at the MSU’s 

Clarksville Research Center (Clarksville, MI). Potato seed tubers were hand planted early- to 

mid-June as four-plant hills in 1.5 m plots in a randomized complete block design with one to 

three replicates. Late blight-susceptible rows and plots were planted around the perimeter and 

between blocks to promote even late blight distribution in the field. After approximately 60 days, 

all plots were inoculated with a zoospore suspension of late blight at 3x106 spores/mL at the end 

of July or beginning of August of each year. Over the 7-year period, different isolates were used 

to infect the trial depending on the prominent isolate in the region. The P. infestans genotype 

(clonal lineage) detected in each year in the trial can be found in Table S2.1. Following 

inoculation, plots were rated visually for the percentage of foliar area affected by late blight. 

Ratings were taken at 3 to 7-day intervals, based on the rate of disease progression for 35-50 

days post inoculation (DPI). Finally, the percent defoliation data was used to calculate the 

relative area under the disease progress curve - RAUDPC (Fry 1978). 

Common scab field resistance trials (370 genotypes and a total of 3,885 disease records) 

were conducted under field conditions during nine years (2009-2017) in a disease nursery at the 

MSU’s Montcalm Research Center (Lakeview, MI). The field was inoculated with common scab 

from aggressive Michigan isolates, and has been cultivated for high disease pressure for the past 

nine years. The trials were planted in a randomized complete block design consisting of one to 

four replications of five-hill plots. After harvesting, mature tubers in plots were assessed for their 

overall plot disease rating scale of 0-5. The rating was based on a combined score for common 

scab coverage and lesion severity in which a rating of 0 indicates zero infection and 3.0 or 

 

47 

greater scores represent highly susceptible genotypes with >50% infection and severe pitted 

lesions (Driscoll et al. 2009). 

SNP genotypes were obtained using the Infinium 8303 Potato Array. Plant DNA was isolated 

from young potato leaves or tubers using the Qiagen DNeasy Plant Mini Kit (Qiagen, Germany), 

following manufacturer’s instructions. DNA was quantified using the Quant-iT PicoGreen 

dsDNA Assay kit (Invitrogen, San Diego, CA). Genotype scoring was performed using the 

GenomeStudio software (Illumina, San Diego, CA). The tetraploid SNP calling was performed 

as per Hirsch et al. (2013), using a custom tetraploid genotype calling based on theta values from 

the Illumina GenomeStudio (Illumina, San Diego, CA) and subsequently filtered, removing poor 

quality markers. SNPs were coded by counting the number of copies of a reference allele (e.g., 

B) where 0 denotes fully homozygous allele (AAAA), 1-3 heterozygous alleles (AAAB, AABB, 

ABBB, respectively) and 4, homozygous allele (BBBB). Finally, the genotype file was filtered 

by retaining SNPs with minor allele frequency (MAF) >0.05 and missing rate <0.15. The 

remaining missing SNP-based genotypes were imputed with the SNP means. The final number 

of SNPs that passed the quality filtering were 4,110. 

We compared the observed and expected rates of heterozygous loci, the later derived under 

the assumption of Hardy-Weinberg (HW) equilibrium. Averaged across loci, the observed rates 

of heterozygosity (0.663) was only slightly higher than the one predicted from estimated allele 

frequencies (0.647). The regression of the observed and expected frequency of heterozygous loci 

had an estimated slope of 0.98 (SE=0.0025) and a R2 of 0.974. Moreover, we did not identify 

any clear outlier SNP that may have indicated a systematic deviation of the observed frequency 

of heterozygous relative to the one predicted from the estimated allele frequency of the locus. 

 

48 

Genomic  relationships  were  computed  from  centered  and  scaled  SNP-based  genotypes 

according  to  VanRaden  (2008): !"#= %&%&'
()*+(%&).  Here, !"#  is  a  matrix  describing  genomic 
relationships between genotypes, %&={[012−4567(012)]/:;(012)} is a matrix of centered and 
scaled SNP-based genotypes (012∈{0,1,2,3,4} counts the number of copies of the reference allele 
at the mth loci. Subtracting the 4567(012), centers the SNP-based genotypes to a null mean and 
dividing by the SNP standard deviation, :;(012), scale SNP-based genotypes to unit variance). 
Finally, division by the number of SNP-based genotypes, 7DEF(%&), makes the average diagonal 
value of !"# equal to one. We use this matrix to quantify genomic relationships and to derive 

principal components, the later were computed by applying the eigen() R-function to GRM. 

ii. 

Statistical analyses 

We use whole-genome regression models (Meuwissen et al. 2001; de los Campos et al. 2013) for 

G1HI=J+∑

QORS

estimation of marker effects and variance component analyses and for assessment of prediction 

accuracy. The general form of the statistical model used was as follows: 

MNO1PO
+TH+U1+U51H+V1HI          [1] 
where G1HI is a phenotypic score (for either late blight or common scab) of the kth replicate of the 
ith genotype collected in year j, J is an intercept, ∑
QORS
SNP-derived principal components, TH are year effects, U1 is the main effect of the ith genotype 
(alternative specifications of this effect are discussed below), U51H represents a genotype-by-year 
interaction and V1HI are error terms, which were treated as normal, independently and identically 
distributed (iid) with year-specific variances, that is V1HIWW;~YZ0,[H\]. 

MNO1PO

 is a regression on the first five 

 

49 

Year had seven levels for late blight and nine levels for common scab (2009, 2010,.., 2017) and 

was treated as a random effect. Genetic and genetic-by-year interactions were also modeled as 

random effects. We considered four specifications for modeling the main effect of genotypes: 

- Genotype effect. In this specification we assumed that the main effects of the genotypes 

where IID draws from normal distributions U1WW;~YZ0,[^\]. In this specification, no genetic 

information (SNPs) was used and no assumptions about gene action (additive, dominance, 

epistasis) were made. This specification was used as a baseline for a model that could be fitted 

without having genomic information. The other three specifications included genotypes as 

inputs. 

0a12b2

cSSd2RS

- Additive + Dominance (A+D). In this case, the main effects of genotypes have an additive 

- Additive model (A).	 Here, the main effect of the genotype was represented using a linear 
  where 0a12=[012−
combination  of  the  marker  genotypes,  that  is U1=∑
4567(012)]/:;(012) were centered and scaled genotypes code at the mth SNP in the ith genotype 
and b2 is the additive effect of the markers. 
0a12b2
component plus one that accounted for dominance; therefore in this model U1=∑
+
cSSd2RS
  where e&12=[e12−4567(e12)]/:;(e12)  are  (centered  and  standardized) 
e&12;2
	∑
cSSd2RS
dummy  variables  for  heterozygous  loci,  here e12=1  (=0)  indicates  that  the  mth  SNP  of  the  ith 
genotype was in heterozygous (homozygous) state and ;2 is the dominant effect of the markers. 
- General model (G). Here, U1=∑
cSSd2RS
standardized) dummy variables for genotypes carrying n copies of the reference allele and P2(.is 

, where f&12( are (centered and 

∑ f&12(P2(
c(Rd

the general effect of the markers. Since there are up to five distinct genotypes (0,1,…,4) this 

model includes up to four degree of freedom per locus. This parameterization allows for any 

 

50 

form of interactions of alleles within locus; thus, it can be considered the most general 

specification for a model accounting for additive and dominance effects. 

a. Prior distributions for effects 

Marker effects (including both additive, dominance and those of the G model) were treated as 

random. We considered two prior distributions of effects: (i) treating SNP effects as draws from 

normal distributions with null mean and model-specific variances (i.e., there were separate 

variances for additive and dominance), this approach was implemented using the Bayesian Ridge 

Regression (BRR) specification in the Bayesian Generalized Linear Regression (BGLR) R-

package (Pérez and de los Campos 2014), and (ii) a Bayesian shrinkage-variable selection 

method (BayesB, Meuwissen et al. 2001). As with BRR, in BayesB different regularization 

parameters (probabilities of non-null effects and scale parameters) were assigned to effects in 

the model argument of the linear predictor. 

additive and dominance. BayesB was implemented in BGLR using the “BayesB” keyword for 

Genotype-by-year effects (U51H) were treated as IID normal with mean zero and variance 
common to all the interactions, that is, U51HWW;~YZ0,[^g\]. 

Sequence of models. Using the specifications described above, we produced a sequence of 

models designed to quantify the amount of variance explained (and the contribution to prediction 

accuracy) of each of the terms entering in the model of expression [1]. The sequence of models 

considered is summarized in Table 2.1. 

 

 

 

 

 

51 

Table 2.1. Sequence of models. 

Model # 
(label)a 

M1 
M2  
M3  
M4 
M5 (A) 
M6 
(A+D) 
M7 (G) 

Effects Included 

 
 
 

Year  Genotypeb  PCc  Additived  Dominancee  Generalf  Genotype-
´ 
´ 
´ 
´ 
´ 
´ 
´ 

by -Yearg  Error 
 
´ 
 
´ 
´ 
´ 
´ 
´ 
´ 

 
 
 
´ 
´ 
´ 
´ 

 
´ 
´ 
´ 
 
 

´ 
´ 
´ 
´ 
´ 

 
´ 
´ 
  

 
 
 
 
 

´ 

  

  

 
 
 

 

 

 

´ 

a M1-M7 are model numbers. b Random effect of the genotype (no SNPs used, no assumption about gene action are 
made). c Principal components, d linear regression on allele content (0/1/2/3/4), e Simple dominance (1 degree of 
freedom per locus representing heterozygous) and f General model for additive + dominance (with up to 4 degrees of 
freedom per locus). g Genotype-by-year interaction. An ‘´’ indicates that the effects was included in the model. 
We  used  the  whole-genome  regression  models  described  above  for  three  purposes:  (1)  estimation  of  variance 
components, (2) identification of variants with high contribution to additive variance and (3) assessment of prediction 
accuracy in cross-validation. 
 

b. Variance components. 

The amount of variance accounted for by each of the terms included in the model was estimated 

using the methods described in de los Campos et al. (2015) and Lehermeier et al. (2017). We use 

these methods to decompose the total phenotypic variance into components due to year, genetics 

factors, genotype-by-year interactions (G´E) and within-year error variance. We also use this 

approach to assess the relative contribution of SNP-additive and dominance effects. 

 

c. Identification of SNPs with a sizable contribution to additive variance. 

Response to selection is directly proportional to additive variance (Falconer and Mackay 1996). 

Thus, in GS, the single-loci additive variance represents a natural metric to assess the relative 

importance of individual loci from a breeding perspective. Under linkage equilibrium, the 

contribution of each locus to additive variance is given by h6i(012b2)=h6i(012)b2\ . In our 
case, genotypes were standardized to unit variance; therefore, h6i(012b2)=b2\ . We used 

 

52 

samples from the posterior distribution of SNP effects from the A model to assess the 

contribution of individual loci to additive variance. 

 

d. Prediction accuracy evaluation.  

We implemented two cross-validation schemes. First, we used a 5-fold cross-validation, 

assigning genotypes to folds. When using this approach all the phenotypic records of a genotype 

are assigned to either training or testing populations. Thus, this approach yields an estimate of 

the prediction accuracy that can be achieved predicting the performance of genotypes that have 

not been evaluated in field trials (i.e., prediction based on genotype data only) and is equivalent 

to the method labeled as Cross-Validation one (CV1) in Burgueño et al. (2012). For this scheme, 

genotypes were assigned to folds completely at random and the 5-fold Cross Validation (CV) 

was repeated 100 times to obtain accurate estimates of the average prediction correlation and its 

standard deviation. 

In a second prediction scheme (CV2), we assigned years to folds (i.e., there as many folds as 

years). Thus, when analyzing the jth fold, data from the jth year was assigned to testing and data 

from all the other years was used for training. This CV approach yields an estimate of the 

prediction accuracy that can be achieved when attempting to predict future year performance 

based on past data. Note that in this case, unlike CV1, when predicting data for the ith genotype 

on the jth year all the data from the ith genotype collected in other years was part of the training 

dataset. In both CV schemes prediction accuracy was evaluated by computing the within-year 

correlation between phenotypes and CV predictions. 

 

 

 

53 

e. Software. 

All the analyses were conducted using R (The R Development Core Team 2010). Models were fit 

using the BGLR-R package. For each model, we ran the Gibbs sampler algorithm for a total of 

33,000 cycles, discarding the first 3,000 samples for burn-in; one of every five of the remaining 

samples was saved and used to estimate posterior means and standard deviations. 

 

IV. Results 

The distribution of late blight and common scab infection varied substantially between years 

(Figure 2.1). In general, RAUDPC median values decreased from 2010 to 2012, with US-22 as 

the prevalent late blight genotype on infected plants. In subsequent years, a differential response 

for late blight resistance was observed when US-23 was the prevalent genotype. Disease pressure 

changes, together with the environment fluctuations between years contribute to explain the 

phenotypic variation observed for the late blight resistance response. Similarly, for common 

scab, a reduced frequency of resistant genotypes (0-1 score) was observed from 2009 until 2013, 

having at the same time an increasing number of intermediate susceptible genotypes (2-3 scores). 

Since 2013 and until 2017, an increased frequency of common scab resistant genotypes was 

observed (Figure 2.1). 

 

 

 

 

54 

 

Figure 2.1. (a) Boxplot of late blight scores (relative area under the disease progress curve- 

RAUDPC), and (b) bar plot for common scab scores (0-5 rating scale). 

 

A principal component (PC) analysis showed that potato genotypes clustered in two groups, one 

involving 391 genotypes, and a small one including 22 genotypes (Figure 2.2). The eigenvalues 

associated to the first two PCs explained about 8% of the total genotype variance (Figure 2.2). 

Figure 2.2. Principal component analysis of the Michigan State University’s potato breeding 

genotypes derived from 4,110 SNPs: loadings on the first two marker-derived principal 

components (a) and proportion of variance explained by the top 10 principal components (b). 

 

 

55 

A cluster analysis using a correlation matrix derived from SNP markers supports the PC-analysis 

results (Figure 2.3). The heatmap also reveals that the strength of genomic relationships among 

the  different  materials  is  relatively  small  (the  clear  majority  of  the  genotypes  have  genomic 

relationships  with  other  genotypes  smaller  than  0.1,  with  only  a  few  genotypes  showing 

relationships comparable to parent-offspring or full-sib relations, i.e., 0.5, Figure 2.3). 

Figure 2.3. Heatmap of the genomic relationship matrix (GRM) from the Michigan State 

 

University’s potato breeding genotypes. 

 

i. 

Variance Components Estimates 

The variance components analyses for late blight resistance (Tables 2.2 and S2.2) revealed that 

year explained roughly 25% of the variance in disease scores. For this trait, and taking as a 

reference the model M3, the main effect of genotype explained about 34% of the variance, 

 

56 

genotype-by-year interactions explained 25% of the variance and the error term explained 

roughly 14% of the variance in late blight scores. These results suggest that a substantial 

proportion of within-year variance in late blight scores (roughly 70%, computed as 

0.34/[0.34+0.144]) can be explained by main effects of genotypes. For late blight, the amount of 

genetic variance captured by the A model was roughly 94% of the variance captured by the G 

model (computed as 0.330/0.352). 

 

Table 2.2. Variance components estimates (posterior standard deviation) derived from BayesB 

model for late blight and common scab resistance by model. Phenotypic scores were standardized 

to unit variance; hence estimates can be interpreted as the proportion of var variance explained by 

each component. Results obtained with the fully Gaussian model (BRR) are presented in Table 

S2.2. 

Model 

# 

(label)a 

Year 

Genotype 

PCb 

Genetic 

Marker-derived 

Additivec  Dominanced  Generale 

0.266 
(0.021) 
0.256 
(0.014) 
0.250 
(0.027) 
0.244 
(0.026) 
0.240 
(0.027) 
0.240 
(0.027) 
0.249 
(0.028) 

0.033 
(0.006) 
0.030 
(0.004) 
0.029 
(0.005) 

 

0.434 
(0.018) 
0.340 
(0.031) 
0.265 
(0.028) 

 

 

  

 

0.456 
(0.016) 
0.440 
(0.021) 

M1 

M2 

M3  

M4 

M5 (A) 
M6 
(A+D) 
M7 (G) 

M1 

M2 

M3 
 

 

Late blight 

 

 

 

 

 

 

0.096 
(0.024) 
0.135 
(0.069) 
0.135 
(0.062) 
0.107 
(0.043) 

 

0.292 
(0.051) 
0.166 
(0.063) 
  

 

 

 

 

  

 

0.141 
(0.049) 

Common Scab 

 

 

 

 

 

 

 

 

 

57 

 

 

 

 

 

 

 

 

 

0.280 
(0.034) 

Total 
geneticf 

 

0.434 
(0.018) 
0.340 
(0.031) 
0.351 
(0.032) 
0.330 
(0.035) 
0.340 
(0.034) 
0.352 
(0.033) 

 

0.456 
(0.016) 
0.440 
(0.021) 

Genotype-
by-yearg 

Error 

 

 

 

0.253 
(0.019) 
0.251 
(0.018) 
0.275 
(0.020) 
0.267 
(0.020) 
0.251 
(0.019) 

 

0.059 
(0.009) 

0.735 
(0.025) 
0.303 
(0.011) 
0.144 
(0.006) 
0.144 
(0.006) 
0.144 
(0.006) 
0.144 
(0.006) 
0.144 
(0.006) 

0.971 
(0.022) 
0.523 
(0.012) 
0.483 
(0.013) 

Table 2.2. (cont’d)  
M4 

 

 

 

 

 

 

0.419 
(0.021) 

M5 (A) 

M6 
(A+D) 

0.029 
(0.006) 
0.031 
(0.006) 
0.031 
(0.006) 
0.031 
(0.006) 

0.030 
(0.012) 
0.132 
(0.087) 
0.107 
(0.073) 
0.051 
(0.030) 

0.485 
(0.013) 
0.485 
(0.013) 
0.485 
(0.013) 
0.484 
M7 (G) 
(0.013) 
a M1-M7 are model numbers (label). The effects included in each of them are described in the columns. b Principal 
components, c linear regression on allele content (0/1/2/3/4), d Simple dominance (1 degree of freedom per locus 
representing heterozygous) and e General model for additive + dominance (with up to 4 degrees of freedom per locus). f 
Total genetic variance, g Genotype-by-year interaction. 

0.447 
(0.023) 
0.443 
(0.025) 
0.448 
(0.024) 
0.451 
(0.023) 

0.056 
(0.009) 
0.061 
(0.009) 
0.057 
(0.009) 
0.056 
(0.009) 

0.507 
(0.076) 
0.356 
(0.087) 

 

0.151 
-0.06 

 

0.442 
(0.031) 

  

  

  

 

For common scab (Tables 2.2 and S2.2) the main effect of genotype explained about 44% of the 

total variance, year and genotype-by-year effects explained only 3% and 6% of the total variance, 

respectively, and the error term accounted for almost one half (48%) of the variance in disease 

scores. For common scab we also observed that the amount of genetic variance captured by the A 

model was very similar to the one captured with the G model. The proportion of the total genetic 

variance that could be attributed to the first-5 PCs was substantial for late blight (~30%, computed 

as 0.107/0.352) and low for common scab (~10%, 0.051/0.451). 

For the A model (fitted using BayesB), we computed single-locus variances and used these 

estimates as proxies for the SNP relevance (Figures 2.4a and 2.4b). Additionally, we report in 

Figures S2.1 and S2.2, linkage disequilibrium (LD) plots for the 10 leading SNPs (i.e., those with 

the larger single-SNP variance) for each trait. For both pathogens, there were a few regions with 

large  single-SNP-variance.  Specifically,  for  late  blight,  there  were  multiple  SNPs  distributed 

across  the  potato  chromosomes  (Figure  2.4a  and  Table  S2.3)  with  a  sizable  contribution  to 

variance, suggesting that multiple genes contribute to the resistant phenotype. Conversely, for 

common scab, there was one SNP, located in chromosome IX (Figure 2.4b and Table S2.4), that 

 

58 

stands  out  for  its  contribution  to  variance  and  a  few  SNPs  with  a  moderate  contribution  to 

phenotypic variance. 

Figure 2.4. Estimated SNP-variances derived from BayesB model using the additive model for 

late blight (a) and common scab (b). (In both cases, phenotypes where diseases scores 

standardize to a variance equal to one. Vertical lines indicate the positions of the top-10, 

according to estimated SNP-variance markers). 

 

 

59 

ii. 

Cross-validation analysis 

The results from the first cross-validation analysis (CV1) yielded an estimated prediction 

correlation of about 0.31 for late blight resistance using the G model. For this trait, there was a 

relatively small, albeit significant, increase in prediction correlation for the G model relative to the 

A model. Likewise, there was a slight superiority of BayesB over BRR (Table 2.3). In the case of 

common  scab,  the  A  model  (with  a  prediction  correlation  of  ~0.27)  outperformed  the  A+D 

(correlation  ~0.26)  and  G  (correlation  ~0.22)  models.  These  results  agree  with  the  variance 

component analyses results, where we also found evidence of a slightly higher relevance of non-

additive effects in the case of late blight, compared to common scab. 

 

Table 2.3. Cross-validation correlations obtained with BRR and BayesB models by trait and 

model. 

Priora 

Model # 
(label) b 

CV-Correlation 

Averagec 

SDd 

M5 (A) 

Proportion of times that the model in row gave a higher 

correlation than the model in columns 
BayesBa 
BRRa 
M6 
M6 
(A+D)  M7 (G) 
(A+D)  M7 (G)  M5 (A) 

BRR 

BayesB 

M5 (A) 
M6 
(A+D) 
M7 (G) 
M5 (A) 
M6 
(A+D) 
M7 (G) 

BRR 

M5 (A) 
M6 
(A+D) 
M7 (G) 
BayesB  M5 (A) 

0.258 
0.241 
0.312 
0.26 

0.24 
0.313 

0.268 
0.259 
0.218 
0.278 

0.023 
0.023 
0.017 
0.024 

0.024 

0.017 

0.025 
0.023 
0.022 
0.026 

 
 

 

Late Blight 
 

0.96 

0.04 
1 
0.67 

0.09 

 

1 
0.96 

0.43 

1 
1 
Common scab 
0.81 

 

0.19 
0.01 
0.9 

 

0.01 
0.93 

60 

0 
0 
 
0 

0 

0.5 

0.99 
0.99 

 
1 

0.33 
0.04 
1 
 

0.06 

1 

0.1 
0.07 
0 
 

0.91 
0.57 
1 
0.94 

 

1 

0.55 
0.2 
0.02 
0.91 

0 
0 
0.5 
0 

0 

  

1 
0.99 
0.63 
1 

Table 2.3. (cont’d) 

 

M6 
(A+D) 
M7 (G) 

0.265 

0.216 

0.025 

0.022 

0.45 

0 

0.8 

0.01 

0.98 

0.37 

0.09 

0 

 

0.02 

0.98 

  

a BRR uses a Gaussian prior for effects, BayesB uses a prior that has a point of mass at zero and a scaled-t slab. b A: 
Additive model, A+D: additive+dominance; G: general model (with up to 4 degrees of freedom per locus).c Average 
from 100 cross-validations. d Standard deviation. 
 

Note that in Table 2.3 we only included results from models using genotypes. Results from other 

models  (e.g.,  M2  and  M3)  are  not  presented  because  in  CV1  they  render  zero  within-year 

correlation.  This  happens  because  in  CV1  predictions  are  entirely  depending  on  borrowing  of 

information between genotypes, a feature that is not possible in models that do not use genotype 

or pedigree information. 

The results from the second cross-validation (i.e., where years were assigned to folds, CV2) 

yielded higher estimates of prediction accuracy than those obtained in CV1 (Tables 2.4 and S2.5). 

This happens because in CV2 there is within-genotype borrowing of information across years. For 

late blight, prediction correlations ranged from 0.41 to 0.74, depending on the model and year. 

Likewise, for common scab, we obtained correlations ranging from 0.46 to 0.76. For both traits, 

the  across-year  average  correlations  showed  small  differences  between  models  (with  a  slight 

superiority in favor of the G model). 

 

Table 2.4. Year cross-validation correlations obtained with BayesB model by trait and model. 

Year 

2010 
2011 
2012 
2013 
2014 
2015 
2017 

M2 

0.551 
0.652 
0.583 
0.422 
0.621 
0.719 
0.508 

 

 

Model # (label) a 

M5 (A) 

M6 (A+D) 

M7 (G) 

M3 

Late blight 

0.537 
0.658 
0.604 
0.415 
0.596 
0.73 
0.504 

0.463 
0.608 
0.586 
0.484 
0.633 
0.678 
0.471 

0.465 
0.611 
0.586 
0.485 
0.64 
0.696 
0.491 

0.517 
0.642 
0.624 
0.492 
0.655 
0.745 
0.506 

61 

Table 2.4. (cont’d) 
Average 

SD 

2009 
2010 
2011 
2012 
2013 
2014 
2015 
2016 
2017 

Average 

SD 

0.579 
0.098 

0.459 
0.52 
0.61 
0.625 
0.75 
0.615 
0.649 
0.639 
0.508 
0.597 
0.088 

0.578 
0.104 

0.56 
0.087 

Common scab 

0.46 
0.522 
0.611 
0.628 
0.759 
0.611 
0.653 
0.639 
0.51 
0.599 
0.09 

0.472 
0.535 
0.622 
0.628 
0.731 
0.634 
0.666 
0.652 
0.519 
0.606 
0.082 

0.568 
0.089 

0.471 
0.533 
0.625 
0.626 
0.737 
0.636 
0.671 
0.647 
0.52 
0.607 
0.083 

0.597 
0.095 

0.466 
0.517 
0.618 
0.634 
0.75 
0.635 
0.659 
0.647 
0.515 
0.605 
0.089 

aM2  includes  year  and  genotype  (no  SNP  information);  M3:  extends  M2  with  the  addition  of  genotype-by-year 
interaction; M5 includes year, first 5 marker-derived PCs, additive effect of SNPs and genotype-by-year interaction; 
M6 expands M5 by adding the effects of dominance; M7 includes year, 5-PCs, genotype-by-year interactions and 
SNPs with up to 4 degrees of freedom per locus (‘General’ model). 
 

V. Discussion 

Genomic selection has been quickly adopted for breeding in diploid species (Heffner et al. 2009; 

Daetwyler et al. 2013; de los Campos et al. 2013). However, the volume of research and the 

adoption of the GS technology for breeding of polypoid species has been much more limited 

(e.g. Habyarimana et al. 2017; Sverrisdóttir et al. 2017). In this study, we demonstrate how 

genomic models commonly used in GS of diploid organisms can be applied for the analysis and 

prediction of disease susceptibility in autotetraploid potato. 

Our results indicate that a sizable fraction of the within-year inter-individual differences in 

disease resistance (about 0.46±0.04 for late blight and 0.45±0.02 for common scab) can be 

explained using 4,110 codominant SNPs from the Infinium 8303 Potato Array used in this study. 

These moderately high genomic heritability estimates for complex disease phenotypes indicates 

that, in principle, genomic prediction could be used successfully to select for resistance to late 

blight and common scab. 

 

62 

Previous studies have reported heritability estimates for these traits; however, differences in 

the nature of the genetic materials (diploid versus tetraploid, hybrids versus genotypes) and of 

the environmental conditions (natural versus induced infection) makes the comparisons across 

studies difficult (Nelson 1978; Braun, Gevens, et al. 2017). For instance, Haynes and Christ 

(1999) reported much higher heritability estimates for late blight resistance (0.8), but this study 

was based on diploid hybrids. For the same trait, estimates of heritability obtained using 

tetraploid genotypes are closer to the ones reported here (ranging from 0.31 to 0.69, Pajerowska-

Mukhtar et al. 2009; Solano et al. 2014). 

For common scab, previous heritability estimates are also highly variable, depending on the 

genetic material and the environmental conditions. For instance, using diploid potatoes derived 

from a cross between wild relatives (S. phureja ´ S. stenotonum) and cultivated potatoes (di-

haploid S. tuberosum ´ S. chacoense), Haynes et al. (2009) and Braun et al. (2017b) reported 

broad sense heritability estimates ranging from 0.18 to 0.72 for different environments. 

However, studies involving tetraploid potatoes have reported higher heritability estimates with 

values ranging from 0.32 to 0.93 (Haynes et al. 1997; Bradshaw et al. 2008; Tai et al. 2009). 

More recently, 18 dedicated common scab and standard breeding program trials were conducted 

in fields with high disease pressure. The broad sense heritability estimates reported from these 

studies ranged from 0.75 to 0.90 for dedicated common scab trials and from 0.06 to 0.82 for 

standard breeding programs trials involving advanced breeding materials (Navarro et al. 2015). 

The amount of variance in disease resistance that could be attributed to genotype-by-year 

interactions was high for late blight and very small for common scab. These differences are 

likely to be due to the different nature and characteristics of infection on the fields used to 

evaluate late blight and common scab. Specifically, for late blight, the mean scores varied 

 

63 

substantially between years (e.g., it was clearly low in 2013) reflecting changes on the late blight 

aggressiveness and late blight genotypes present in different years, resulting in a large extent of 

genotype-by-year interactions for this pathogen. On the other hand, our common scab data was 

generated in a nursery that has been used to evaluate common scab resistance in potato breeding 

genotypes for several years. Consequently, there was less variability in the mean scores across 

years and therefore we observed substantially less extent of G´E. A similar result was reported 

under comparable conditions by Murphy et al. (1995). Results based on fields trials performed in 

different locations for this pathogen have shown much higher variability over the years (Haynes 

et al. 2009). 

The comparison of the genomic variance estimates obtained with the A model and those 

obtained with the G model suggest that, for both pathogens, a sizable fraction of the total genetic 

variance (0.94 and 0.98, for late blight and common scab, respectively) can be captured by an 

additive model (Table 2.2 and Table S2.2). The amount of genetic variance captured by the A 

model reflects an estimate of the variance that can be captured by regression on allele content 

(i.e., by allele substitution effects). However, when dominance is included in the model (A+D), 

the estimated ‘additive variance’ no longer represents the variance explained by allele 

substitution effects; therefore, the additive component in the A+D model is smaller than the 

additive component estimated with the A model. 

While our variance component estimates indicate that most of the genetic variance can be 

captured by an A model, our cross-validation analysis suggests that accounting for non-additive 

effects could improve prediction accuracy by a small but statistically significant margin in the 

late blight case. These results agree with the theory that suggests that dominance and epistasis 

are expected to contribute to the expression of traits subjected to directional selection or those 

 

64 

affecting the plant fitness such as late blight resistance (Killick and Malcolmson 1973). This may 

explain why the G model captured slightly more variance and predicted slightly more accurately 

late blight scores than the A model. 

The presence of linkage disequilibrium (LD) between loci makes the partition of the total 

genetic variance into (orthogonal) locus-specific components not possible (de los Campos et al. 

2015). However, it is worth looking at the relative size of estimated effects to explore features of 

the genetic architecture of the trait. We did this by inspecting the estimated SNPs variances 

(Figure 2.4). Overall, the proportion of variance explained by individual SNPs was low, 

reinforcing the idea that resistance to both common scab and late blight is polygenic. However, 

there were some SNPs with SNP-variances that compared with most of them were large. For late 

blight, these analyses lead to the finding of many relatively large-variance SNPs located mainly 

in chromosomes V and IX (Figure 2.4a and Table S2.3). For late blight resistance, multiple 

quantitative trait loci (QTLs) have been reported across the 12 potato chromosomes in tetraploid 

and diploid potato populations (Tiwari et al. 2013). Most of these major QTLs are located in 

chromosomes III, IV, V, VII, XI and XII, characterized for harboring hotspot regions for 

resistance to late blight and other pathogens, not only for genes involved in quantitative 

resistance such as R genes, but also for genes involved in qualitative resistance (Malosetti et al. 

2007; Pajerowska-Mukhtar et al. 2009; Álvarez et al. 2017). For instance, genes involved in 

carbohydrate metabolism such as sucrose synthase (Table S2.3) play an active role in the defense 

response elicitation. Sucrose synthesis down-regulation has been described in the Capsicum 

annuum - Phytophthora nicotianae pathosystem, showing a decreasing concentration after 

challenging with beta-aminobutyric acid (BABA) or non-host pathogens and priming the 

synthesis of metabolites associated with the production of defense-related compounds (Stamler 

 

65 

et al. 2015). Additionally, these results validate earlier QTL reports obtained from MSU-derived 

populations using potato varieties carrying different late blight resistance genes coming from 

species previously used in resistance breeding such as S. demissum and S. berthaultii (Massa et 

al. 2015; Manrique-Carpintero, personal communication). 

For common scab resistance, our results suggest an additive resistance effect with a clear 

major-effect SNP located on chromosome IX (Figure 2.4 and Table S2.4). This SNP is 

associated with a WRKY transcription factor known for their role in the modulation of the 

resistance responses in systemic and acquired plant resistance, activating or repressing the 

transcription of genes involved in the synthesis of defense related-proteins such as R proteins 

(Pandey and Somssich 2009; Buscaill and Rivas 2014). In addition to the loci discussed above, 

we were also able to identify additional SNPs with a sizable contribution to variance across the 

potato chromosomes (Figure 2.4). Interestingly, the SNP in the WRKY gene that appeared to 

have a sizable contribution to inter-individual differences in common scab resistance is located 

in a region where LD is relatively weak (see Figure S2.2). 

For instance, we found one SNP located in chromosome III (Table S2.4) associated with the 

primary metabolism-related protein fructokinase, whose concentration increases under pathogen 

attack as a mechanism to reduce the costs attributed to the defense response in soil-borne 

pathogens (Zimaro et al. 2011). Likewise, in chromosome V, we found one SNP related to the 

RNA synthesis-related protein DEAD-box ATP-dependent RNA helicase, reported for its role in 

plant resistance by enhancing the defense response in both necrotrophic and biotrophic 

pathogens (Li et al. 2008). Overall, the evidence we found support the hypothesis that resistance 

to common scab involves multiples mechanisms of defense including the activation of genes 

related to systemic and R gene-mediated resistance. 

 

66 

There are few studies reporting QTLs for common scab resistance. For instance, two QTL 

located in chromosome XI were detected in a diploid parental-derived population for the 

percentage of surface area infected and lesion type caused by common scab, explaining 21% and 

18.2% of the total phenotypic variance, respectively (Braun, Endelman, et al. 2017). For 

tetraploid populations, Amplified Fragment Length Polymorphisms (AFLPs) and Simple 

Sequence repeats (SSRs) markers have been used to establish an association between potato 

genotypes and the common scab resistance phenotype in a tetraploid bi-parental derived-

population. Two copies of a dominant allele were detected in a QTL localized in chromosome II, 

explaining 8.1 and 7.1% of the phenotypic variance, respectively. A second QTL was localized 

in chromosome VI explaining 6.9 % of the total phenotypic variance (Bradshaw et al. 2008). 

Therefore, the large-variance SNP detected in this study represents a new genomic region 

associated with common scab resistance, providing a framework for the development of 

molecular markers for marker-assisted selection and understand the genetics behind common 

scab resistance. 

Our variance component estimates suggest that for both, late blight and common scab, a 

sizable amount of inter-individual differences in disease resistance can be captured using whole-

genome regressions. However, the successful implementation of GS requires being able to 

predict future outcomes from past data. We assessed this using two CV analyses. Our results are 

based on genotypes derived from potato breeding programs. Some of these genotypes are related 

through pedigrees and there is some level of population stratification. Therefore, the prediction 

accuracies reported in our study should be considered representative of the prediction accuracy 

that one may be able to achieve when applying GS to breeding populations. 

We considered two different prediction problems and implemented different CV schemes to 

 

67 

represent each prediction problem. Our first CV focused on the prediction of future scores from 

genotypes that were not evaluated in field trials (i.e., prediction based on information from other 

genotypes). These analyses rendered moderately low CV-correlations (~0.22-0.31 with some 

small differences between traits and models). 

It is important to highlight that in CV1 the correlations reflect the prediction accuracy that can 

be achieved when predicting future phenotypes for genotypes that have not been evaluated in 

field trials. These predictions, although imperfect, could enable several rounds of rapid selection 

based on genotype data alone. The predictive correlation obtained in CV1 was about half of the 

correlation between phenotypes across years (compare results in Table 2.3 with those for M2 in 

Table 2.4). Thus, we conclude that with the array and sample size used in this study, the 

predictive accuracy for late blight and common scab scores obtained from a newly developed 

genotype that has been genotyped but not tested in the field is about half of the predictive power 

of a single phenotype record. If more than two selection cycles can be carried out per year, the 

reduction on generation interval that can be achieved with genomic prediction would overcome 

the lower accuracy and, eventually lead to faster yearly genetic gains. 

Our second CV used years as folds; therefore, in this case, disease scores predictions for one-

year data were obtained from the same genotypes over years. The results of the model based on 

year and genotype (M2), give a baseline estimates of the prediction accuracy that can be 

achieved with phenotypic prediction. In CV2, we obtained higher prediction correlations (0.56-

0.61, Table 2.4) than with CV1. However, the performance of the genomic models was only 

slightly superior to predictions based on past phenotypes-only (i.e., those that could be obtained 

with the M2 model). This result agrees with previous studies (e.g., Crossa et al. 2010) that show 

that the benefits of genomic prediction are more important when predicting phenotypes of 

 

68 

materials that have no (or very limited) data from previous field trials. 

We confirmed that a sizable fraction of inter-individual differences in late blight and common 

scab scores can be attributed to genetic factors and can be captured using whole-genome 

regressions. We found large genotype-by-year interactions for late blight and limited genotype-

by-year interactions for common scab. For both late blight and common scab, we found that an 

additive model could account for a sizable (>90%) of the total genetic variance. However, for 

late blight, we found small (but statistically significant) gains in prediction accuracy when 

accounting for dominance. Our analyses confirm strong associations with disease resistance to 

SNPs in previously reported resistance hotspot regions for late blight and reported a novel locus 

that has a sizable contribution to common scab resistance. We demonstrated that prediction of 

disease resistance, using genomic prediction applied to autotetraploid potato, is feasible and can 

be implemented for SNP-based selection in potato breeding. Further research is needed to 

explore ways (larger sample size, more controlled environments, higher marker density) in which 

genomic prediction accuracy can be further improved.  

 

 

69 

APPENDIX 

 

70 

Figure S2.1. Linkage disequilibrium plots for the top ten ranked loci according to their contribution to the late blight genotypic 

variance. The blue asterisk shows the SNP position on each plot. 

 

 

 

 

 

71 

Figure S2.2. Linkage disequilibrium plots for the top ten ranked loci according to their contribution to the common scab genotypic 

variance. The blue asterisk shows the SNP position on each plot. 

 

 

 

 

72 

Table S2.1. Late blight clonal lineages detected throughout seven years at the Michigan State 

University Clarksville Research Center (Clarksville, MI). 

Year 
2010 
2011 
2012 
2013 
2014 
2015 
2017 

Clonal 
lineage 
US-22 
US-22 
US-22 
US-23 
US-23 
US-23 
US-23 

73 

 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

Table S2.2. Variance components estimates (posterior standard deviation) derived from BRR model for late blight and common scab 

resistance by model. Phenotypic scores were standardized to unit variance; hence estimates can be interpreted as the proportion of 

variance explained by each component. 

Year 

Genotype 

Genetic 

Marker-derived 

PC b 

Additivec 

Dominanced 

Generale 

Total 
genetic f 

Genotype-
by-year g 

Error 

0.266 (0.021) 
0.257 (0.014)  0.434 (0.018) 
0.249 (0.028)  0.340 (0.031) 
0.247 (0.026)  0.266 (0.029)  0.097 (0.025) 
0.238 (0.028) 
0.242 (0.027) 
0.245 (0.027) 

 
 
  

0.139 (0.069)  0.270 (0.046) 
0.132 (0.060)  0.162 (0.061) 
0.105 (0.038) 

 

 

 
 
 

 
 
 

Late blight 

 
 
 
 

 
 
 
 
 

  
  
Common Scab 
 
 
 
 
 
 
 
 
 

0.735 (0.025) 
0.434 (0.018) 
0.303 (0.011) 
0.340 (0.031)  0.253 (0.019)  0.144 (0.006) 
0.353 (0.033)  0.250 (0.018)  0.144 (0.006) 
0.314 (0.036)  0.285 (0.022)  0.144 (0.006) 
0.329 (0.035)  0.272 (0.020)  0.144 (0.006) 
0.271 (0.033)  0.346 (0.033)  0.253 (0.019)  0.144 (0.006) 

0.13 (0.045) 

 

 

 
 

 
 

 
 
 
 
 
 

 
 
 
 
 
 

Model # 
(label)a 

  

M1 
M2 
M3  
M4 
M5 (A) 
M6 (A+D) 
M7 (G) 

0.033 (0.006) 
0.030 (0.004)  0.456 (0.017) 
0.029 (0.005)  0.440 (0.020) 
0.029 (0.005)  0.419 (0.021)  0.030 (0.012) 
0.032 (0.006) 
0.031 (0.006) 
0.031 (0.006) 

 
 
  

M1 
M2 
M3 
M4 
M5 (A) 
M6 (A+D) 
M7 (G) 
a M1-M7 are model numbers (label). The effects included in each of them are described in the columns. b Principal components, c 
linear regression on allele content (0/1/2/3/4), d Simple dominance (1 degree of freedom per locus representing heterozygous) and e 
General model for additive + dominance (with up to 4 degrees of freedom per locus). f Total genetic variance, g Genotype-by-year 
interaction. 

0.971 (0.022) 
0.456 (0.017) 
0.523 (0.012) 
0.440 (0.020)  0.060 (0.008)  0.483 (0.013) 
0.447 (0.023)  0.056 (0.009)  0.485 (0.012) 
0.443 (0.025)  0.061 (0.009)  0.485 (0.013) 
0.445 (0.025)  0.057 (0.009)  0.486 (0.013) 
0.441 (0.031)  0.448 (0.023)  0.056 (0.009)  0.485 (0.013) 

0.117 (0.076)  0.499 (0.070) 
0.079 (0.052)  0.278 (0.085)  0.205 (0.073) 
0.054 (0.031) 

  

  

 

 

74 

Table S2.3. Top ten ranked loci from the whole genome regression derived from BayesB model, 

according to their additive variance estimates for late blight. 

Locus 

Chromosome 

Proportion of variance 

explained (Prob.)* 

Putative function 

60S ribosomal protein L34 

Pre-mRNA-splicing factor syf2 

Ammonium transporter 1 member 1 

Sucrose synthase 

Gene of unknown function 

6-phosphogluconate dehydrogenase, 
decarboxylating 

Chalcone synthase 

Dihydropterin pyrophosphokinase-
dihydropteroate synthase 

Xanthine dehydrogenase 

HCF106; proton motive force 
dependent protein transmembrane 
transporter  

* Estimated posterior probability of having a SNP variance greater than zero given the data. 

solcap_snp_c2_23056 

solcap_snp_c1_11991 

solcap_snp_c2_20640 

solcap_snp_c2_27764 

solcap_snp_c2_50301 

solcap_snp_c2_33537 

solcap_snp_c2_54368 

solcap_snp_c2_23829 

solcap_snp_c2_14946 

solcap_snp_c1_4444 

V 

X 

IX 

IX 

V 

V 

V 

V 

XI 

III 

0.00441 
(0.669) 
0.00227 
(0.596) 
0.00264 
(0.590) 
0.00262 
(0.578) 
0.00143 
(0.490) 
0.00121 
(0.475) 
0.00109 
(0.467) 
0.00098 
(0.427) 
0.00083 
(0.449) 
0.00066 
(0.409) 

 

  

 

75 

Table S2.4. Top ten ranked loci from the whole genome regression derived from BayesB model, 

according to their additive variance estimates for common scab. 

Locus 

Chromosome 

Proportion of variance 

explained (Prob.)* 

Putative function 

0.0306 
(0.977) 
0.0019 
(0.695) 
0.0018 
(0.766) 
0.0015 
(0.690) 
0.0014 
(0.698) 
0.0011 
(0.678) 
0.001 
(0.699) 
0.001 
(0.683) 
0.0009 
(0.643) 
0.0007 
(0.409) 

WRKY transcription factor 

Fructokinase 

TAF5 

Conserved gene of unknown 
function 

DEAD-box ATP-dependent RNA 
helicase 22 

Auxin response factor 

Leucine-rich repeat-containing 
protein 

Homocysteine s-methyltransferase 

Dead box ATP-dependent RNA 
helicase 

solcap_snp_c2_34103 

Conserved gene of unknown 
function 
* Estimated posterior probability of having a SNP variance greater than zero given the data. 

solcap_snp_c2_54867 

solcap_snp_c2_42312 

solcap_snp_c1_8133 

solcap_snp_c1_15485 

solcap_snp_c2_45522 

solcap_snp_c2_52104 

solcap_snp_c1_2519 

solcap_snp_c2_22003 

solcap_snp_c2_6185 

IX 

III 

VI 

VII 

V 

VII 

I 

IX 

XI 

VIII 

 

 

 

 
76 

Table S2.5. Year cross-validation correlations obtained with BRR model by trait and model. 

Year 

2010 
2011 
2012 
2013 
2014 
2015 
2017 

Average 

SD 

2009 
2010 
2011 
2012 
2013 
2014 
2015 
2016 
2017 

Average 

SD 

M2 

0.551 
0.651 
0.583 
0.424 
0.621 
0.72 
0.509 
0.58 
0.097 

0.458 
0.52 
0.61 
0.625 
0.751 
0.615 
0.649 
0.639 
0.507 
0.597 
0.088 

Model # (label) a 

M3 

M5 (A) 

M6 (A+D) 

M7 (G) 

Late blight 

0.531 
0.661 
0.611 
0.419 
0.597 
0.724 
0.504 
0.578 
0.102 

0.456 
0.605 
0.574 
0.482 
0.634 
0.683 
0.467 
0.557 
0.09 

Common scab 

0.461 
0.521 
0.613 
0.629 
0.758 
0.612 
0.654 
0.639 
0.51 
0.6 
0.09 

0.469 
0.534 
0.627 
0.626 
0.734 
0.63 
0.67 
0.65 
0.517 
0.606 
0.083 

0.463 
0.612 
0.581 
0.492 
0.654 
0.692 
0.491 
0.569 
0.089 

0.471 
0.534 
0.628 
0.626 
0.736 
0.634 
0.672 
0.646 
0.518 
0.607 
0.083 

0.518 
0.643 
0.614 
0.49 
0.647 
0.735 
0.499 
0.592 
0.092 

0.466 
0.519 
0.618 
0.634 
0.75 
0.636 
0.66 
0.647 
0.514 
0.605 
0.089 

aM2 includes year and genotype (no SNP information); M3: extends M2 with the addition of genotype-by-year 
interaction; M5 includes year, first 5 marker-derived PCs, additive effect of SNPs and genotype-by-year interaction; 
M6 expands M5 by adding the effects of dominance; M7 includes year, 5-PCs, genotype-by-year interactions and 
SNPs with dominance modeled using up to 4 degrees of freedom per locus (‘General’ model). 
 

 

 

 
77 

LITERATURE CITED 

 

 
78 

LITERATURE CITED 

 
 

 
Ahn, Y. K., and T.-H. Park. (2013). Resistance to common scab developed by somatic hybrids 

between Solanum brevidens and Solanum tuberosum. Acta Agriculturae Scandinavica, 
Section B - Soil & Plant Science 63: 595–603. 

Álvarez, M. F., M. Angarita, M. C. Delgado, C. García, J. Jiménez-Gomez et al. (2017). 
Identification of novel associations of candidate genes with resistance to late blight in 
Solanum tuberosum Group Phureja. Frontiers in Plant Science 8: 1040. 

Barone, A. (2004). Molecular marker-assisted selection for potato breeding. American Journal of 

Potato Research 81: 111–117. 

Bradshaw, J. E., C. A. Hackett, B. Pande, R. Waugh, and G. J. Bryan. (2008). QTL mapping of 

yield, agronomic and quality traits in tetraploid potato (Solanum tuberosum subsp. 
tuberosum). Theoretical and Applied Genetics 116: 193–211. 

Braun, S. R., J. B. Endelman, K. G. Haynes, and S. H. Jansky. (2017). Quantitative trait loci for 

resistance to common scab and cold-induced sweetening in diploid potato. The Plant 
Genome 10: 1–9. 

Braun, S., A. Gevens, A. Charkowski, C. Allen, and S. Jansky. (2017). potato common scab: a 

review of the causal pathogens, management practices, varietal resistance screening methods, 
and host resistance. American Journal of Potato Research. 94(4): 283–296. 

Burgueño, J., G. D. L. Campos, K. Weigel, and J. Crossa. (2012). Genomic prediction of 

breeding values when modeling genotype × environment interaction using pedigree and 
dense molecular markers. Crop Science 52: 707–719. 

Buscaill, P., and S. Rivas. (2014). Transcriptional control of plant defense responses. Current 

Opinion in Plant Biology 20: 35–46. 

Crossa, J., G. de los Campos, P. Pérez, D. Gianola, J. Burgueño et al. (2010). Prediction of 

genetic values of quantitative traits in plant breeding using pedigree and molecular markers. 
Genetics 186: 713 LP-724. 

Daetwyler, H. D., M. P. L. Calus, R. Pong-Wong, G. de los Campos, and J. M. Hickey. (2013). 

Genomic prediction in animals and plants: simulation of data, validation, reporting, and 
benchmarking. Genetics 193: 347 LP-365. 

Dees, M. W., and L. A. Wanner. (2012). In search of better management of potato common scab. 

Potato Research 55: 249–268. 

 

 
79 

Drewnowski, A., and C. D. Rehm. (2013). Vegetable cost metrics show that potatoes and beans 

provide most nutrients per penny. PLoS One 8(5): e63277. 

Driscoll, J., J. Coombs, R. Hammerschmidt, W. Kirk, L. Wanner et al. (2009). Greenhouse and 

field nursery evaluation for potato common scab tolerance in a tetraploid population. 
American Journal of Potato Research 86: 96–101. 

Falconer, D., and T. Mackay. (1996). Introduction to quantitative genetics (4e). Pearson United 

Kingdom. 

FAOSTAT. (2018). FAOSTAT database collections. Food and Agriculture Organization of the 

United Nations. 

Fry, W. E. (1978). Quantification of general resistance of potato cultivars and fungicide effects 

for integrated control of potato late blight. Phytopathology 68: 1650–1655. 

Gebhardt, C., A. Ballvora, B. Walkemeier, P. Oberhagemann, and K. Schüler. (2004). Assessing 
genetic potential in germplasm collections of crop plants by marker-trait association: A case 
study for potatoes with quantitative variation of resistance to late blight and maturity type. 
Molecular Breeding 13: 93–102. 

Habyarimana, E., B. Parisi, and G. Mandolino. (2017). Genomic prediction for yields, processing 

and nutritional quality traits in cultivated potato (Solanum tuberosum L.). Plant Breeding 
136: 245–252. 

Haverkort, A. J., P. C. Struik, R. G. F. Visser, and E. Jacobsen. (2009). Applied biotechnology to 
combat late blight in potato caused by Phytophthora infestans. Potato Research 52: 249–264. 

Haynes, K. G., and B. J. Christ. (1999). Heritability of resistance to foliar late blight in a diploid 

hybrid potato population of Solanum phureja × Solanum stenotomum. Plant Breeding 118: 
431–434. 

Haynes, K. G., B. J. Christ, C. R. Burkhart, and B. T. Vinyard. (2009). Heritability of resistance 

to common scab in diploid potatoes. American Journal of Potato Research 86: 165–170. 

Haynes, K. G., R. W. Goth, and R. J. Young. (1997). Genotype × environment interactions for 

resistance to common scab in tetraploid potato. Crop Science 37: 1163–1167. 

Heffner, E. L., M. E. Sorrells, and J.-L. Jannink. (2009). Genomic selection for crop 

improvement. Crop Science 49: 1–12. 

Hirsch, C. N., C. D. Hirsch, K. Felcher, J. Coombs, D. Zarka et al. (2013). Retrospective view of 

north american potato (Solanum tuberosum L.) breeding in the 20th and 21st centuries. G3: 
Genes, Genomes, Genetics 3: 1003–1013. 

 

 
80 

Killick, R. J., and J. F. Malcolmson. (1973). Inheritance in potatoes of field resistance to late 

blight [Phytophthora infestans (Mont.) de Bary]. Physiological Plant Pathology 3: 121–131. 

Kolasa, K. (1993). The potato and human nutrition. American Potato Journal 70: 375–384. 

Lehermeier, C., G. de los Campos, V. Wimmer, and C.-C. Schön. (2017). Genomic variance 
estimates: With or without disequilibrium covariances? Journal of Animal Breeding and 
Genetics 134: 232–241. 

Li, D., H. Liu, H. Zhang, X. Wang, and F. Song. (2008). OsBIRH1, a DEAD-box RNA helicase 

with functions in modulating defense responses against pathogen infection and oxidative 
stress. Journal of Experimental Botany 59: 2133–2146. 

de los Campos, G., J. M. Hickey, R. Pong-Wong, H. D. Daetwyler, and M. P. L. Calus. (2013). 

Whole-genome regression and prediction methods applied to plant and animal breeding. 
Genetics 193: 327 LP-345. 

de los Campos, G., D. Sorensen, and D. Gianola. (2015). Genomic heritability: What is it?. 

PLOS Genetics 11(5): e1005048. 

Lozano, R., O. Ponce, M. Ramirez, N. Mostajo, and G. Orjeda. (2012). Genome-wide 

identification and mapping of NBS-encoding resistance genes in Solanum tuberosum Group 
Phureja. PLoS One 7(4): e34775. 

Malosetti, M., C. G. Van Der Linden, B. Vosman, and F. A. Van Eeuwijk. (2007). A mixed-
model approach to association mapping using pedigree information with an illustration of 
resistance to Phytophthora infestans in potato. Genetics 175: 879–889. 

Massa, A. N., N. C. Manrique-Carpintero, J. J. Coombs, D. G. Zarka, A. E. Boone et al. (2015). 

Genetic linkage mapping of economically important traits in cultivated tetraploid potato 
(Solanum tuberosum L.). G3: Genes, Genomes, Genetics 5: 2357–2364. 

Meuwissen, T. H. E., B. J. Hayes, and M. E. Goddard. (2001). Prediction of total genetic value 

using genome-wide dense marker maps. Genetics 157: 1819–1829. 

Mosquera, T., M. F. Alvarez, J. M. Jiménez-Gómez, A. Draffehn, A. Hofmann et al. (2016). 

Targeted and untargeted approaches unravel novel candidate genes and diagnostic SNPs for 
quantitative resistance of the potato (Solanum tuberosum L.) to Phytophthora infestans 
causing the late blight disease. PLoS One 11(6): e0156254. 

Muktar, M. S., J. Lübeck, J. Strahwald, and C. Gebhardt. (2015). Selection and validation of 

potato candidate genes for maturity corrected resistance to Phytophthora infestans based on 
differential expression combined with SNP association and linkage mapping. Frontiers in 
Genetics 6: 1–19. 

 

 
81 

Murphy, A. M., H. De Jong, and G. C. C. Tai. (1995). Transmission of resistance to common 

scab from the diploid to the tetraploid level via 4x-2x crosses in potatoes. Euphytica 82: 227–
233. 

Navarro, F. M., K. T. Rak, E. Banks, B. D. Bowen, C. Higgins et al. (2015). Strategies for 
selecting stable common scab resistant clones in a potato breeding program. American 
Journal of Potato Research 92: 326–338. 

Nelson, R. R. (1978). Genetics of horizontal resistance to plant diseases. Annual Review of 

Phytopathology 16: 359–378. 

Nowicki, M., M. R. Foolad, M. Nowakowska, and E. U. Kozik. (2011). Potato and tomato late 

blight caused by Phytophthora infestans: An overview of pathology and resistance breeding. 
Plant Disease 96: 4–17. 

Pajerowska-Mukhtar, K., B. Stich, U. Achenbach, A. Ballvora, J. Lübeck et al. (2009). Single 
Nucleotide Polymorphisms in the Allene Oxide Synthase 2 Gene are associated with field 
resistance to late blight in populations of tetraploid potato cultivars. Genetics 181: 1115 LP-
1127. 

Pandey, S. P., and I. E. Somssich. (2009). The Role of WRKY Transcription factors in plant 

immunity. Plant Physiology 150: 1648–1655. 

Pérez, P., and G. de los Campos. (2014). Genome-wide regression & prediction with the BGLR 

Statistical Package. Genetics 198(2): 483–495 

Pomerantz, A., Y. Cohen, E. Shufan, Y. Ben-Naim, S. Mordechai et al. (2014). Characterization 

of Phytophthora infestans resistance to mefenoxam using FTIR spectroscopy. Journal of 
Photochemistry and Photobiology B: Biology 141: 308–314. 

Solano, J., I. Acuña, F. Esnault, and P. Brabant. (2014). Resistance to Phytophthora infestans in 

Solanum tuberosum landraces in Southern Chile. Tropical Plant Pathology 39: 307–315. 

Sparks, A. H., G. A. Forbes, R. J. Hijmans, and K. A. Garrett. (2014). Climate change may have 

limited effect on global risk of potato late blight. Global Change Biology 20: 3621–3631. 

Stamler, R. A., O. Holguin, B. Dungan, T. Schaub, S. Sanogo et al. (2015). BABA and 
Phytophthora nicotianae induce resistance to Phytophthora capsici in chile pepper 
(Capsicum annuum). PloS One 10: 1–19. 

Sverrisdóttir, E., S. Byrne, E. H. R. Sundmark, H. Ø. Johnsen, H. G. Kirk et al. (2017). Genomic 

prediction of starch content and chipping quality in tetraploid potato using genotyping-by-
sequencing. Theoretical and Applied Genetics 130: 2091–2108. 

 

 
82 

Tai, G. C. C., A. M. Murphy, and X. Xiong. (2009). Investigation of long-term field experiments 
on response of breeding lines to common scab in a potato breeding program. Euphytica 167: 
69–76. 

The Potato Genome Sequencing Consortium. (2011). Genome sequence and analysis of the tuber 

crop potato. Nature 475: 189–195. 

The R Development Core Team (2010). R : A Language and Environment for Statistical 

Computing. Retrieved from http://www.R-project.org/ 

Tiwari, J. K., S. Siddappa, B. P. Singh, S. K. Kaushik, S. K. Chakrabarti et al., (2013). Molecular 

markers for late blight resistance breeding of potato: an update. Plant Breeding 132: 237–
245. 

VanRaden, P. M. (2008). Efficient methods to compute genomic predictions. Journal of Dairy 

Science 91: 4414–4423. 

Wanner, L. A. (2006). A survey of genetic variation in Streptomyces isolates causing potato 

common scab in the United States. Phytopathology 96: 1363–1371. 

Wanner, L. A., and W. W. Kirk. (2015). Streptomyces – from basic microbiology to role as a 

plant pathogen. American Journal of Potato Research 92: 236–242. 

Zimaro, T., N. Gottig, B. S. Garavaglia, C. Gehring, and J. Ottado. (2011). Unraveling plant 

responses to bacterial pathogens through proteomics. Journal of Biomedicine and 
Biotechnology 2011: 354801. 

 

 

 

 

 

 

 

 

 

 

 

 

 
83 

OVERCOMING SELF-INCOMPATIBILITY IN DIPLOID POTATO USING CRISPR-

Chapter 3  

CAS9 

[Front. Plant Sci. doi: 10.3389/fpls.2019.00376] 

 

Felix Enciso-Rodriguez1, Norma C. Manrique-Carpintero1, Satya Swathi Nadakuduti1, C. Robin 

Buell2,3,4, Daniel Zarka1, David Douches.1,4* 

 

1 Department of Plant, Soil and Microbial Sciences, Michigan State University, East Lansing, 

Michigan, 48824. 2 Department of Plant Biology, Michigan State University, East Lansing, 

Michigan, 48824. 3 Plant Resilience Institute, Michigan State University, East Lansing, 

Michigan. 4 Michigan State University AgBioResearch, East Lansing, Michigan. 

* Corresponding author (douchesd@msu.edu) 

 

 

 

 
84 

I. Abstract 

Potato breeding can be redirected to a diploid inbred/F1 hybrid variety breeding strategy if self-

compatibility can be introduced into diploid germplasm. However, the majority of diploid potato 

clones (Solanum spp.) possess gametophytic self-incompatibility (SI) that is primarily controlled 

by a single multiallelic locus called the S-locus which is composed of tightly linked genes, S-

RNase (S-locus RNase) and multiple SLFs (S-locus F-box proteins), which are expressed in the 

style and pollen, respectively. Using S-RNase genes known to function in the Solanaceae 

gametophytic SI mechanism, we identified S-RNase alleles with flower-specific expression in 

two diploid self-incompatible potato lines using genome resequencing data. Consistent with the 

location of the SLF gene in potato, we genetically mapped the S-RNase gene using a segregating 

population to a region of low recombination within the pericentromere of chromosome I. To 

generate self-compatible diploid potato lines, a dual single-guide RNA (sgRNA) strategy was 

used to target conserved exonic regions of the S-RNase gene and generate targeted knock-outs 

using a Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR-associated 

approach. Self-compatibility was achieved in nine S-RNase knock-out (KO) T0 lines which 

contained bi-allelic and homozygous deletions/insertions in both genotypes, transmitting SC to 

T1 progeny. This study demonstrates an efficient approach to achieve stable, consistent self-

compatibility through S-RNase KO for use in diploid potato breeding approaches. 

 

II. Introduction 

Cultivated potato (Solanum tuberosum L.) is the third most important food crop after rice and 

wheat (Devaux et al. 2014) and plays an essential role in human nutrition as a primary source of 

carbohydrates. Although global production of potato totaled 388 million tons in 2017 

 

 
85 

(FAOSTAT 2019), potatoes face high production losses due to biotic and abiotic stresses that 

will increase with global warming (Raymundo et al. 2017). While improvement of cultivated 

potatoes (2n=4x=48) relies on the discovery and introgression of genes from wild species for 

traits such as disease resistance, the polyploid nature of cultivated tetraploid potato hampers the 

fixation of desirable alleles in new cultivars. For example, the introgression of critical dominant 

alleles such as the potato virus Y (PVY) disease-resistance gene in a triplex or quadruplex allelic 

configuration can take up to 15 years (Mendoza et al. 1996). Re-inventing potato as a diploid 

inbred/F1 hybrid variety (2n=2x=24) would allow the application of efficient breeding methods 

(Jansky et al. 2016) as inbred potatoes would accelerate the generation of new varieties with 

favorable allelic combinations targeting yield, tuber quality, and resistance traits. A significant 

barrier to this approach is the occurrence of gametophytic self-incompatibility (SI) in a majority 

of the diploid potato germplasm, thereby preventing the ability to generate diploid homozygous 

lines. 

In diploid potato, the gametophytic SI system is controlled by a single multiallelic locus 

called the S-locus (Porcher and Lande 2005). This locus is composed of tightly linked genes, S-

RNase (S-locus RNase) and SLFs (S-locus F-box) genes known also as S-haplotype-specific F-

box brothers (SFBB), expressed in the style and pollen, respectively (Takayama and Isogai 2005; 

Sassa et al. 2007; Kubo et al. 2010; Bush and Moore 2012). The S-RNase protein produces 

cytotoxic effects that inhibit the elongation of self-pollen tubes via degradation of RNA from the 

pollen whereas SLF function as a component of a detoxification complex that mediates 

ubiquitination of non-self S-RNase proteins leading to degradation via the proteasome pathway 

(Sijacic et al. 2004; Kubo et al. 2015). Hence, when self-pollination occurs in self-incompatible 

 

 
86 

individuals, the SLFs genes do not recognize their native S-RNase and consequently, pollen tube 

growth in the style is inhibited due to the ribonuclease activity of the S-RNase (Hua et al. 2008). 

In an effort to develop diploid-self-compatible (SC) potato lines, the inbred line M6 was 

generated from the wild tuber-bearing species, Solanum chacoense (Jansky et al. 2014). In M6, a 

dominant allele of the S-locus inhibitor (Sli) inactivates the gametophytic SI system (Hosaka and 

Hanneman 1998) leading to self-compatibility. However, introgression of Sli into other 

germplasm is time-consuming and could lead to linkage drag and fixation of undesirable traits 

such as high tuber glycoalkaloid content from the donor S. chacoense. An alternative strategy to 

Sli introgression is the use of genome editing to accelerate the generation of SC diploid lines by 

targeting genes involved in SI. 

Genome editing by Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR)-

associated protein 9 (Cas9) system has been widely used to generate gene knock-outs (KO) of 

candidate genes related to agronomic traits in important crops (Jaganathan et al. 2018). Cas9 

induces double-strand breaks (DSB) in the DNA at the target site, triggering the response of 

endogenous cell repair mechanisms. One of the cellular mechanisms to repair DSBs is non-

homologous end joining (NHEJ), which can generate insertions and deletions in the coding 

region resulting in a KO of gene function (Bortesi and Fischer 2015; Pellagatti et al. 2015). The 

target-DNA recognition is mediated by a single guide-RNA (sgRNA) bearing a 20 bp target-site 

complementary to region adjacent to a protospacer-adjacent motif (PAM), 5′-NGG-3’, resulting 

in the generation of a DSB (Doudna and Charpentier, 2014; Pellagatti et al., 2015; Sternberg et 

al., 2014) which can be leveraged to generate DSB of target genes. 

Previous studies in tomato wild relatives demonstrated that missense mutations and gene loss 

prevent S-RNase ribonuclease activity in S. peruvianum and S. pennellii, leading to self-

 

 
87 

compatibility (Royo et al. 1994; Kowyama et al. 1994; Covey et al. 2010; Li and Chetelat 2015). 

Considering that S-RNase is the gametophytic SI component directly implicated in degradation 

of RNA in self-pollen tubes, inhibiting the S-RNase function is a straightforward strategy to 

confer SC in potato. In an effort to contribute to the development of diploid inbred potato lines, 

we generated SC diploid lines by targeted mutagenesis of S-RNase using CRISPR-Cas9, 

obtaining stable self-compatibility in T0 and T1 generations. Contemporaneous with the writing 

of this manuscript, Ye et al. (2018) published their findings using a similar approach. However, 

this study provides further insight into SI in diploid potatoes, reporting three new S-RNase 

alleles, their localization within a low recombination pericentromeric region consistent with the 

location of the S-locus, generation of stable SC in KO lines, and documentation of plasticity in 

the phenotype of SI in two diploid lines. 

 

III. Material and Methods 

i. 

Plant material 

After an initial test of self-compatibility with more than 50 self-pollinations, the SI diploid 

potato lines (2n=2x=24) DRH-195 and DRH-310 F1 lines were generated from a cross between 

S. tuberosum Gp. Phureja DM 1-3 516 R44 (DM) and S. tuberosum Gp Tuberosum RH 89-039-

16 (RH) at Virginia Tech and used in this study. Plants were maintained in vitro, propagated on 

Murashige and Skoog (MS) medium (MS basal salts plus vitamins, 3% sucrose, 0.7% plant agar, 

pH 5.8) (Murashige and Skoog, 1962) and cultured in growth chambers with 16-h-light/8-h-dark 

photoperiod at 22°C and average light intensity of 200 μmoles m-2s-1. 

 

 

 
88 

ii. 

Allelic identification, annotation and phylogenetic analysis of S-RNase 

TBLASTN (BLAST- basic local alignment search tool) searches were performed using 

reported S-RNase protein sequences (Table 3.1) from the Solanaceae family against the DM 

v4.04 assembly (Hardigan et al. 2016) using BLAST v2.2.31 (Altschul et al. 1990) with default 

parameters. A candidate S-RNase gene was selected using the top blast hits. Expression 

abundances across a range of developmental stages, tissues, and organs were determined using 

available gene expression atlases for DM and RH (The Potato Genome Sequencing Consortium, 

2011). To identify S-RNase allelic variants in the diploid potato clones, genomic and 

complementary DNA sequence data from DRH-195 and DRH-310 leaf and tuber tissues were 

retrieved from the Sequence Read Archive (SRA) of the National Center for Biotechnology 

Information (Table S3.1) and aligned to the DM v4.04 assembly using BWA-MEM (Li 2013). 

Duplicate reads were removed using Picard Tools v1.113 (http://broadinstitute.github.io/picard) 

and consensus sequences were obtained using the mpileup utility from Samtools v1.2 (Li et al. 

2009) with the consensus option from bcftools v1.2 (Li 2011). 

A primer set was designed to amplify the predicted ORF of the S-RNase gene in DRH-195 

and DRH-310 using the detected S-RNase variants (Table S3.2). S-RNase amino acid sequences 

reported in Table 3.1, including the alleles reported by Ye et al. (2018), along with the deduced 

amino acid sequences from the S-RNase variants identified in this study were aligned using 

Clustal Omega (Sievers et al. 2011). A phylogenetic tree was constructed using the Neighbor 

joining method with 1000 bootstrap replicates in MEGA version 7.0 (Kumar et al. 2016). Amino 

acid similarities percentages were calculated using BioEdit (Hall 1999). 

 
 
 
 

 

 
89 

Table 3.1. S-RNase sequences from seven Solanaceae species used in this study. 

Species 
Solanum tuberosum 
Nicotiana sylvestris 
Solanum neorickii 
Solanum chilense 
Solanum chacoense 
Petunia integrifolia 
subsp. inflata 
Petunia x hybrida 

Gene/protein†, 
Ribonuclease S-2 
RNase 
S-RNase 
S1-RNase 
S11 
S2 self-incompatibility 
ribonuclease precursor 
Sx-protein 

Accession* 
Q01796 
CAA05306 
BAC00940 
BAC00934 
AAA50306 
AAG21384 

AAA33729 

* Gene or protein name, ‡National Center for Biotechnology Information (NCBI) accession identifier. 
 

iii. 

S-RNase linkage mapping 

The previously reported diploid DRH mapping population was used to genetically map the S-

RNase locus (Manrique-Carpintero et al., 2015). DNA was isolated from DRH-195 and DRH-

310 young leaves using the DNeasy Plant Mini Kit (Qiagen, Hilden, Germany), and used for 

PCR with a Q5® High-Fidelity DNA Polymerase (New England Biolabs, Ipswich, MA) with the 

following thermocycler conditions: one cycle of initial denaturation for 4 min at 94 °C, followed 

by 34 cycles for 15 s at 94 °C, 45 s at 56 °C and 1 min at 72 °C and a final extension of 5 min at 

72 °C. S-RNase amplicons were gel-purified using the QIAquick PCR Purification Kit (Qiagen, 

Hilden, Germany) and cloned into the Zero Blunt TOPO PCR Cloning vector (ThermoFisher, 

Carlsbad, CA). Ten colonies for each line were sequenced by the Sanger method and aligned 

using Clustal Omega (Sievers et al. 2011). DM and RH allelic sequences were confirmed and 

used to design S-RNase RH-allele specific primers (Table S3.2). These primers were screened 

across 80 individuals of the DRH mapping population. The genotype from the presence/absence 

of an RH allele was coded as nnxnp and used for mapping in JoinMap4.1 with the same 

parameters as previously reported by Manrique-Carpintero et al. (2015). 

 

 

 
90 

iv. 

sgRNA identification, assembly and validation 

A double sgRNA construct targeting predicted conserved regions from the first (sgRNA 1) and 

second (sgRNA 2) S-RNase exons were designed using CRISPR RGEN tools (Table S3.2, Park 

et al., 2015). A gene KO construct containing the sgRNA combination (sgRNA 1-2) was 

assembled in the pHSE40 vector containing the CRISPR-Cas9 cassette as described by Xing et 

al. (2014) and transferred into Agrobacterium tumefaciens strain GV3101 pMP90 (Koncz et al. 

1994) by electroporation. 

 

v. 

Agrobacterium-mediated transformation 

Agrobacterium-mediated transformation was performed using leaf segments from four-week-old 

tissue culture plants of DRH 195 and DRH 310 as described by Li et al. (1999). Briefly, explants 

were pre-cultured on a step I media (MS salts, 3% sucrose, 5 g/l phytoagar, 1 mg/l thiamine-HCl, 

0. 8 mg/l zeatin-riboside and 2 mg/l 2,4-D) for four days and inoculated with Agrobacterium. 

After three days, explants were rinsed with sterile distilled water containing 250 mg/l cefotaxime 

and 200 mg/l carbenicillin and placed onto step II media (MS salts, 3% sucrose, 5 g/l phytoagar, 

1 mg/l thiamine-HCl, 0.8 mg/l zeatin- riboside, 2 mg/l gibberellic acid, 20 mg/l hygromycin and 

150 mg/l ticarcillin disodium and clavulanate potassium). Explants were transferred to fresh step 

II media every week. After approximately 30 days, transformation events (T0 lines) were 

selected from step II media and transferred to root induction media containing MS medium 

supplemented with antibiotics for selection as described above. 

 

 

 
91 

vi. 

Molecular characterization of KO lines 

DNA from T0 and T1 plants was isolated as described above. PCR was carried out using the 

GoTaq DNA polymerase (Promega, Fitchburg, WI) with the following thermocycler conditions: 

one cycle of initial denaturation for 4 min at 94 °C, followed by 34 cycles for 15 s at 94 °C, 45 s 

at 56 °C and 1 min at 72 °C and a final extension of 5 min at 72 °C. Amplicons were visualized 

on 1% (w/v) agarose gels. Allelic mutations of positive transformation events were identified by 

insertion/deletion presence. Selected transformation events were amplified with the Q5 High-

Fidelity DNA Polymerase (New England Biolabs, Ipswich, MA). Then, purified PCR products 

were cloned into the Zero Blunt TOPO PCR Cloning vector (ThermoFisher, Carlsbad, CA), and 

transformed into DH5a competent cells (ThermoFisher, Carlsbad, CA). Colonies carrying the 

alleles from each event were Sanger sequenced. 

 

vii. 

Assessment of self-compatibility 

One month old in-vitro plants were planted in one gallon plastic pots with a peat and perlite 

growth medium mixture (Bacto professional planting mix) and placed into a greenhouse with a 

light intensity of 250 µmoles m−2 s−1, 16/8-h light/dark photoperiod and a temperature of 25 

°C. Plants were fertilized with Peters Professional® 20: 20: 20 fertilizer (The Scotts Co., 

Marysville, OH) at a rate of 500 mg/l twice a week. Around 50 flowers per plants were hand 

self-pollinated to test for SC. Pollen staining with acetocarmine-glycerol (Ordoñez 2014a) and 

cross-pollination were also done to test male and female viability, respectively. T0 fruits were 

harvested three-four weeks after self-pollination and kept at room temperature for two weeks. 

Extracted T1 seeds were sterilized and subjected to overnight treatment with 1500 ppm of 

gibberellic acid then allowed to germinate. T1 seedlings were transferred to greenhouse and self-

 

 
92 

pollinated as described above. Additionally, chloroplast counting of guard cells was performed 

according to Ordoñez (2014a) to discard possible chromosome doubling in each selected S-

RNase KO line. 

 

viii. 

S-RNase expression analysis 

Twenty-five flowers from wild-type (WT) and DRH-195/310-derived T0 KO lines (DRH-

195.158 and DRH-310.21) were self-pollinated at anthesis. Pollinated pistils were excised 24-

hour post pollination (hpp) and preserved in -80 ºC until use. Total RNA was isolated using the 

RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) and DNase treated using the TURBO DNA-

free kit (ThermoFisher, Carlsbad, CA) following manufacturer’s instructions. RNA was 

quantified using a NanoDrop spectrophotometer (ThermoFisher Scientific, Grand Island, NY) 

and a reverse-transcription polymerase chain reaction (RT-PCR) was carried-out with 1 µg of 

total RNA using the Super-Script One-Step RT-PCR kit (Thermo Fisher Scientific, Carlsbad, 

CA). Primers designed to amplify the S-RNase ORF and the elongation-factor one alpha (EF1a) 

housekeeping internal control were used for the RT-PCR reaction (Table S3.2). 

 

IV. Results 

i. 

Identification of the S-RNase gene in potato 

For this study, we used available genomic and gene expression data from the sequenced doubled 

monoploid DM and the SC heterozygous breeding line RH (The Potato Genome Sequencing 

Consortium 2011). To identify the S-RNase gene, the DM genome sequence was selected as the 

reference genome and utilized in sequence similarity searches using seven Solanaceae S-RNase 

genes (Table 3.1). Candidate genomic regions encoding the S-RNase gene were identified and 

 

 
93 

located within 3,948,850 - 3,949,581 bp of the unanchored scaffold PGSC0003DMB000000091 

and the annotated DM S-RNase allele, PGSC0003DMG400026738, which encodes a 738 bp 

open reading frame and a 216 amino acids (aa) predicted protein composed of five conserved 

and two hypervariable regions, characteristic of SI S-RNases (Figure 3.1A; Ioerger et al., 1991). 

The detected DM and RH S-RNase alleles resembles class III S-RNases (Figure 3.1B) and is 

comprised of two exons and one small intron, which is located at position five of the 11 

recognized intron positions for this gene family (Igic and Kohn 2001; Ramanauskas and Igić 

2017). 

We found that the DM S-RNase (referred hereafter as S. tub_Sp5 from S. tuberosum S-RNase 

allele five of S. tuberosum Group Phureja) is highly expressed in mature flowers (245.5 

Fragments Per Kilobase of exon model per Million mapped reads – FPKM) compared with no 

expression in leaves or tubers in DM. High S-RNase expression levels were also detected in 

carpels (4342.7 FKPM) consistent with its role in preventing SC (Kao and Tsukamoto 2004). 

More limited gene expression data is available for RH (The Potato Genome Sequencing 

Consortium 2011) and consistent with the expression of S-RNase in potato, it is expressed in 

flowers (167.04 FPKM) and not tubers or leaves. Together with the functional annotation, these 

results suggest we have identified the S-RNase gene involved with SI. 

 

 
94 

 

Figure 3.1. S-RNase gene structure and allelic variants in diploid potato and related species. (a) 

S-RNase predicted amino-acid sequence alignment of the DM (Sp5) and RH (St5 and St6) alleles. 

Underlined regions in red represent the typical five conserved regions (C1 to C5) and two 

hypervariable regions (HVa and HVb) of the S-RNase gene family. Exon/intron boundary is 

indicated with a filled triangle within the HVa region. (b) S-RNase gene structure. The S-RNase 

open reading frame is composed of two exons separated by one small intron. Zoomed-in regions 

are shown within dotted lines indicating the intronic and exonic regions used for RH-specific 

primer design within the reported S-RNase alleles. (c) Phylogenetic tree constructed using the 

 

 
95 

Neighbor Joining method based in the proportion of S-RNase amino acid differences. S-RNase 

from N. sylvestris was used as out-group. Numbers above each branch represent bootstrapping 

percentages from1000 replications. (d) Pairwise amino acid similarity of S-RNase in Solanum 

species and detected S-RNase alleles. S-RNase alleles are represented first by a species name 

abbreviation followed by S and the allele number for Tuberosum (t) or Phureja (p) group in S. 

tuberosum. For other species, a similar pattern is used, and allele numbers or letters (i.e. Sx for 

P. hybrida) are added if reported. S. tub: S. tuberosum, S. chc: S. chacoense, S. neo: S. neorickii, 

P. int: P. integrifolia, P. hyb: P. hybrida, N. syl: N. sylvestris. 

 

ii. 

Identification of allelic variants of S-RNase in diploid potato lines 

The DRH-195 and DRH-310 F1 diploid self-incompatible lines derived from a cross between 

DM and RH were used to identify S-RNase allelic variants. Using whole genome resequencing 

data for these two lines, the DM and RH S-RNase alleles were identified in DRH-195 and DRH-

310 and validated using Sanger sequencing. As previously observed for DM, the S. tub_Sp5 

predicted 216 amino acid sequence was detected in both lines, beside one of the two RH S-

RNase alleles in each line, hereafter referred as S. tub_St5 and S. tub_St6 (S. tuberosum S-RNase 

alleles five and six of S. tuberosum Group Tuberosum group respectively, Figure 3.1A). 

A phylogenetic tree was constructed using available S-RNase amino acid sequences from 

Solanum species and the allelic variants identified in this study (Figure 3.1C). Two main clades 

with high confidence bootstrap values were observed. The S. tub_Sp5, S. tub_St5 and S. tub_St6 

alleles and the S. chacoense (S. chc_S11) allele exhibited highest percentage of similarity relative 

to the other species (Table S3.3, Figure 3.1D), whereas the Petunia (P. int_S2 and P. hyb_Sx) and 

S. tuberosum (S. tub_St2) S-RNases clustered in two separate subgroups. Two S. tuberosum 

 

 
96 

Group Phureja alleles, S. tub_Sp3 and S. tub_Sp4, were located in separate clades (Figure 3.1C). 

Overall, the amino acid sequence identity between S-RNase alleles ranged from 42 to 100%, 

showing 92% similarity between the DM and RH S-RNase alleles and 93% between the two RH 

alleles (Figure 3.1D). Notably, the S. tub_Sp4 allele had 100% similarity with a previously 

reported S-RNase from S. tub_St2 (NCBI accession: Q01796), presumably representing the same 

allele. Moreover, S. tub_Sp5, S. tub_St5, and S. tub_St6 alleles had similar amino acid identity (53-

54%) to the reported S. tub_Sp3 allele when compared with S. tub_Sp4 (Ye et al. 2018). 

 

iii. 

S-RNase is located within a pericentromeric region of chromosome I in potato 

Using the segregating DRH population, linkage mapping indicated that the S-RNase S. tub_St6 

allele mapped to the pericentromeric region of chromosome I spanning a region between 13.7 

and 17.8 cM (solcap_snp_c2_27882 and solcap_snp_c1_16425 markers, respectively), 

corresponding to 6.1 Mb and 18.9 Mb of chromosome I in the physical map (Figure 3.2). These 

results are consistent with the region corresponding to the map location of the S-locus in potato 

(Gebhardt et al. 1991). 

 

 

 

 

 

 

 
97 

 

Figure 3.2. S-RNase gene mapping in diploid potato. (a) RH S-RNase (S. tub_St6) allelic 

screening on the DRH F1 population using the S-RNase and housekeeping Sucrose synthase 3 

gene (Sus-3) primers. DM and RH parental lines are shown in the first two lanes followed by the 

negative control [C], and the RH-S-RNase segregation pattern of 11 F1-derived lines. (b) The S-

RNase gene mapped to 16.3 cM on the short arm near the centromeric region of chromosome I 

(red). (c) Marey map of physical (Mb) versus genetic (cM) distances from chromosome I 

 

 
98 

showing the S-RNase gene within a low-recombination region (red box). Asterisks within the red 

box represent SNPs spanning the region between to 6.1Mb and 18.9 Mb in the potato physical 

map (solcap_snp_c2_27882 and solcap_snp_c1_16425 markers, respectively). 

 

iv. 

Targeted mutagenesis of S-RNase in diploid potatoes using CRISPR/Cas9 results in self 

compatibility 

A dual gRNA strategy (sgRNA 1 and sgRNA 2) was used to efficiently generate S-RNase KOs 

and disrupt the S-RNase function in DRH-195 and DRH-310 (Figure 3.3A). Multiple T0 plants 

were recovered for each line due to a 98% regeneration and transformation efficiency for DRH-

195 and 93% for DRH-310 (Table 3.2). Based upon PCR analysis using primers to the S-RNase 

and gel detection of insertion/deletion polymorphisms, biallelic S-RNase mutations were 

recovered in both the DRH-195 and DRH-310-derived T0 lines (Figure 3.3B, 3.3C). Specifically, 

seven S-RNase KOs exhibiting polymorphic deletions with up to 580 bp were detected for DRH-

195-derived T0 lines. In contrast, for DRH-310, only three S-RNase KOs were detected with up 

to 524 bp monomorphic deletions. To further characterize the CRISPR-targeted regions in 

selected T0 lines, both T0 KOs and WT-like S-RNase amplicons were sequenced (Figure 3.3D, 

3.3E). A distinct nucleotide deletion was detected in each KO line, ranging from small bi-allelic 

deletions (1 bp) to large homozygous deletions (527 bp) in both S-RNase alleles of each DRH-

derived T0 lines. Insertions (1 to 18 bp) and inversions (486 bp) were also observed in a bi-allelic 

configuration. Similarly, besides the described mutation types, chimeric mutations were detected 

in T0 lines, which has been reported in other species subjected to CRISPR-mediated 

mutagenesis, potentially due to late embryogenesis editing (Gomez et al. 2018). 

 

 

 
99 

 

Figure 3.3. CRISPR/Cas9 mediated mutagenesis in two SI diploid potato lines. (a) Single-guide 

RNAs (sgRNAs) designed to target the S-RNase exon1 (sgRNA 1) and exon 2 (sgRNA 2). 

Zoomed-in regions within the dotted lines show sgRNA and PAM sequences. (b) Seven and (c) 

three bi-allelic S-RNase knock-outs (KOs) DRH-195 and DRH-310 T0 lines, respectively, 

detected by insertion/deletion polymorphisms compared to wild type (WT) and negative controls 

[C]. All DRH-195 KOs exhibit polymorphic deletions and DRH-310 monomorphic deletions 

(lines 195-105,195-137,195-142 and 195-160 presented a faint mutated or WT-like bands that is 

not observed in the figure). (d) Different mutation types detected by amplicon sequencing in 

selected KOs for DRH-195, and (e) DRH-310 T0 lines respectively. Wild-type (WT) S-RNase 

 

100 

 

exhibiting selected sgRNAs (sgRNA1 and sgRNA2) and PAM sequences are shown at the top of 

each alignment. Different types of mutations including deletions (-), insertions (+), and 

inversions (i) detected in DM (DM) and RH (RH) S-RNase alleles of each T0 KO DRH-derived 

lines shown as ‘195-’ or ‘310-’. 

To test whether the selected T0 KOs lines underwent spontaneous chromosome doubling, 

chloroplasts were counted in stomatal guard cells. One out of ten S-RNase KO lines (DRH-

195.104) revealed chromosome doubling which has also been observed in a related S-RNase KO 

approach (Ye et al., 2018). This phenomenon, known as endopolyploidization, is frequently 

observed in potatoes subjected to regeneration processes, in which structural cell and 

chromosome rearrangements at mitosis results in increased chromosome numbers (Karp et al. 

1984; Owen et al. 1988). The tetraploid KO line was not considered for further analysis. 

 

Table 3.2. DRH-195 and DRH-310-derived T0 and indel-based selected S-RNase KO lines with 

bi-allelic mutations. 

Line 

Num. 

Explants 

DRH-195 
DRH-310 

186 
276 

T0  
lines 
162 
78 

 *Calculated as the percentage of T0 lines with Cas9 integration 
 

Transformation 
Efficiency (%)* 

98 
93 

Mutant deletion 

polymorphism type 
Single 
Double 

 
3 

7 
 

To confirm the S-RNase mutant phenotype, T0 KO lines were self-pollinated in two separate 

replications under greenhouse conditions. In both replications, all T0 KO lines set fruit (Figure 

3.4A, Tables S3.4 and S3.5). In one of the replications, both wild-type non-transformed lines 

(DRH-195 and DRH-310) also exhibited a limited number of specific self-pollination events 

with fruit set that either had complete development (DRH-195, Table S3.4) or arrest of fruit set 

 

101 

 

two weeks after self-pollination (DRH-310, Figure S3.5A). However, after one-week, new self-

pollinations of wild-type lines did not set fruit (Figure 3.4B) suggesting plasticity of self-

compatibility, a phenomenon observed previously in Solanum (Saba-El-Leil et al. 1994; Mena-

Ali and Stephenson 2007). 

 

Figure 3.4. S-RNase expression and knock-out phenotype in SI diploid potato lines. (a) Fruits 

obtained after five weeks of self-pollination in an S-RNase DRH-195-derived T0 mutant line. (b) 

Dropped flowers after self-pollination in the wild type DRH-195. (c) Semi-quantitative reverse 

transcription PCR (RT-PCR) in DRH-195 and DRH-310 self-pollinated wild-type (WT) and 

knock-out (KO) T0 lines (DRH-195.158 and DRH-310-21, respectively). RNA was isolated from 

 

102 

 

pistils 24 hours after self-pollination revealing S-RNase expression in WT but not in KO lines as 

compared with the housekeeping gene control (EF1a). (d) T1 plants derived from the DRH-

195.158 T0 line were screened with the S-RNase primers. WT-like bands with 1 bp deletion on 

each target site (causing frameshift leading to a premature stop codon, Figure S3.1) are observed 

in T0 and T1 lines. Cas9 gene did not transmit to T1 line 5 (lane 5). A previously undetected band 

observed in lane 6 is potentially the result of transgenerational CRISPR/Cas9 activity. T0: DRH-

195.158, C: Negative control. The red box is showing a T1 line segregating out Cas9 while 

maintaining the S-RNase KO. 

This plasticity however, represents an unreliable source of self-compatibility as was evident 

in the ratio of fruit set per pollination observed, with self-pollination success in S-RNase KO 

lines being an order of magnitude higher than the WT in both DRH-195 and DRH-310 KO lines 

(Tables S3.4 and S3.5). To further investigate if this result was associated with the suppression 

of S-RNase expression, a semi-quantitative RT-PCR was performed. As shown in Figure 3.4C, 

S-RNase transcripts were detected in both wild-type lines 24 hpp yet no expression was detected 

in T0 KOs, confirming S-RNase expression in WT but not mutant lines. 

 

 

 

 

 

 

103 

 

 

Figure 3.5. Fruit formation in WT and T1 S-RNase KO diploid potatoes lines. (a). Fruit setting 

arrest two weeks after self-pollination in the wild type DRH-310. (b). Fruits obtained after four 

weeks of self-pollination in an S-RNase DRH-195-derived T1 mutant line. 

 

 

Viable T1 seeds were obtained for each S-RNase T0 KO line. Self-compatibility was confirmed in 

T1 lines after self-pollination so far, demonstrating the inheritance and stability of the S-RNase 

KO phenotype (Figure 3.5B). Cas9 inheritance in the T1 lines exhibited a segregation ratio 

associated with a hemizygous multi-copy integration of Cas9 (4 out of 135), in addition to the 

segregation of the mutated S-RNase alleles (Figure 3.4D). Likewise, because of the activity of 

 

104 

 

integrated Cas9, a potential transgenerational deletion was observed in a T1 line (DRH-

195.158.6). These results demonstrate the advantage of using CRISPR/Cas9 to generate Cas9-

free edited plants and the potential to transmit stable gene mutations through different 

generations. 

 

V. Discussion 

Self-incompatibility has been a limiting factor for inbred/F1 hybrid cultivar development in 

diploid potatoes because efforts involving crossing with wild SC relatives result in many 

undesirable traits segregating in the progeny. To redirect potato breeding toward an efficient 

inbred/F1 hybrid generation strategy, we exploited the S-RNase-based SI system in diploid 

potatoes and generated KO lines using CRISPR-based genome editing to achieve self-

compatibility. 

Amino acid sequence variation within S-RNase was observed among S. tub_Sp5, S. tub_St5, 

and S. tub_St6 alleles (Figure 3.1A). Nearly half of these variants were within the hypervariable 

domains (HVa and HVb) and not in conserved domains (C1-5) consistent with data that show the 

S-RNase variable regions are the determinants for allele specificity in different Solanum species 

(Matton et al. 1997, 1999; Brisolara-Corrêa et al. 2015). Specifically, four amino acids within 

these variable regions (T74, N76, Y77, and R101) have been reported as the sole factors for 

allele conversion of the pollen rejection phenotype in S. chacoense (Matton et al. 1997). Three of 

these amino acid changes were present within the S. tub_Sp5, S. tub_St5, and S. tub_St6 alleles 

indicating that these variations could be sufficient to confer allele specificity while preserving 

their catalytic activity which is associated with two of the five conserved domains (Kao and 

Tsukamoto, 2004). 

 

105 

 

Ioerger et al. (1990) observed that S-RNase inter-specific similarities were higher than intra-

specific similarities in Solanaceae, concluding that S-RNase divergence pre-dates speciation in 

this clade. The results observed in our study further confirm this previous observation. A high 

degree of inter-specific S-RNase amino acid sequence similarity was observed in Solanum S-

RNases (S. tuberosum and S. chacoense). Conversely, a clear intra-specific separation within the 

S. tuberosum S-RNase alleles (Figure 3.1C) was also observed, consistent with the hypothesis of 

a single ancestral origin of S-RNase and conservation of specific polymorphisms throughout 

evolution governing allelic diversity (Ioerger et al. 1991; Dzidzienyo et al. 2016). 

The S-RNase gene mapped to chromosome I within a region of low recombination consistent 

with the hypothesis to promote outbreeding due to a reduction in recombination events between 

the S-RNase and SLF genes (Kubo et al. 2015; Fujii et al. 2016). This chromosome position has 

also been reported in other Solanaceae members. For instance, S-RNase is located on 

chromosome I in S. lycopersicum and S. peruvianum within highly complex and repetitive 

genomic regions (Kubo et al. 2015; Fujii et al. 2016).  Furthermore, Kubo et al. (2015) mapped 

an SLF, the other component of the S-locus, also to chromosome I in potato genome within a 

repeat-rich sub-centromeric region, suggesting that the S-RNase location was at the same 

position since these genes are reported to be closely linked (Sijacic et al., 2004). 

All edited T0 lines had a frameshift in the coding region close to sgRNA 1, leading to a 

premature stop codon. The resulting truncated sequence prevented the amplification of the S-

RNase gene by removing the primer annealing site at the 3’ end (Figure S3.1). Similarly, the 

consistent mutations generated by the two sgRNAs allowed detection of S-RNase size 

polymorphisms. It should be noted that this strategy was selected for the potential to use PCR for 

a quick and facile screen for large deletions in T0 lines. However, undetected insertions/deletions 

 

106 

 

or inversions could be present in T0 lines. For instance, sequencing data revealed a single bp 

insertion and deletion in the DM S-RNase allele of DRH-195.128 and DRH-195.158, 

respectively, showing a similar amplicon size in agarose gels as WT (Figure 3.3B, 3.3D). These 

observations indicate that a large number of allelic KOs can be generated given the high 

transformation efficiency observed in both diploid lines. 

The DRH-195.158 T0 KO line, which exhibited a single bp deletion at each sgRNA targeting 

site in the RH S. tub_Sp5 allele (Figure 3.3D), showed a new S-RNase deletion in a T1-derived 

line (Figure 3.4D). Given the Cas9 mismatching tolerance, this allele possibly underwent a new 

mutagenesis event, displaying a different mutation pattern in the T1 generation. In different plant 

species, it has been found that editing occurs at a higher frequency across generations, therefore 

new mutations segregate from WT alleles in heterozygous T0 as a result of constitutive 

expression of Cas9 (Feng et al. 2014; Xu et al. 2015; Wang et al. 2018). 

Two independent self-pollination assays were conducted in DRH-195 and DRH-310 wild-

type lines in 2015 and 2018 with a minimal of 50 flowers, demonstrating their SI nature. 

However, a third biological replicate in 2018 resulted in fruit formation suggesting plasticity in 

the strength of SI. Environmental effects along with plant phenology have been associated with 

unstable SI in angiosperms. For instance, temperature fluctuations, photoperiod, glucose 

starvation and humidity significantly reduced SI in S. peruvianum after selfing (Webb and 

Williams 1988). This process, known as pseudo-self-incompatibility, has also been reported in 

grasses in which artificial self-pollination techniques can contribute to SI breakdown (Do Canto 

et al. 2016). Similarly, sporadic fruit set has been observed across Solanaceae species such as 

Witheringia solanacea, S. carolinense, S. peruvianum and N. alata in which floral age, flowering 

stage and delayed floral abscission has been associated with fruit set in SI populations (Stone et 

 

107 

 

al. 2006; Mena-Ali and Stephenson 2007; Miller and Kostyun 2011; Liao et al. 2016). This 

phenomenon has also been observed in species under sporophytic SI, in which floral age reduces 

the expression of the S-locus associated genes in Brassica oleracea resulting in SI breakdown 

(Hadj-Arab et al. 2010). In natural populations of Campanula rapunculoides, strong SI has been 

observed in young flowers. However, self-fruit formation is also evident in old flowers as a 

consequence of pollen scarcity and low fruit production from prior inflorescences (Stephenson et 

al. 2000). Therefore, environmental conditions favoring SI breakdown (plant age, plant health, 

and greenhouse conditions) could lead to fruit set in one of the wild-type biological replicates in 

this study. 

Unlike pseudo-self-incompatibility, the S-RNase KO proved to be both stable and consistent 

across different replications and generations, presenting a higher ratio of fruit set per pollination 

when compared with self-fruit WT lines (Tables S3.4 and S3.5). Although the SC phenotype 

appears to be line dependent, distinctive S-RNase KO lines exhibited either high fruit set or seed 

formation. These results also indicated that genes other than S-RNase could be contributing to 

the strength of the SI response in both WT and S-RNase KOs. In fact, besides the S-RNase gene, 

other SI modifier loci can modulate the pollen rejection response in several Solanum species 

(Goldraij et al., 2006; McClure et al., 1999; O’Brien et al., 2002). For instance, S-RNase-

independent stylar factors such as eEF1A or High Top-Band (HT-B) proteins, can directly or 

indirectly interact with S-RNase contributing to the SI response (Goldraij et al. 2006; Soulard et 

al. 2014). Similarly, unintended somaclonal variation and chromosomal rearrangements 

associated with the potato regeneration processes and Cas9 activity respectively, could also 

contribute to variations in the observed ratio of fruit set per pollination within the S-RNase KOs. 

 

108 

 

This hypothesis is further supported by Peterson et al. (2016) which identified several 

genomic regions associated with self-fertility in a DRH F1 population, located on chromosomes 

IV, IX, XI and XII. They also found that a specific SNP associated to the RH allele, fixed in 

selfed populations, is likely the primary factor for self-fertility in the DRH F1 progeny. Overall, 

this study demonstrates that S-RNase is the primary component for self-pollen rejection in DRH-

195 and DRH-310. However, external evidence suggests that besides RH self-fertility 

mechanisms, S-RNase-independent stylar factors and environmental conditions could play a role 

in spontaneous self-compatibility observed in the WT lines in this study. 

 

In this study, we generated self-compatible potato diploid lines by targeting the S-RNase 

gene using the CRISPR-cas9 system. We first computationally identified three new S-RNase 

alleles in SI diploid lines (a DM and two RH alleles, each inherited to DRH-195 and DRH-310, 

respectively) and mapped this gene to chromosome I within the peri-centromeric region 

consistent with the localization of the S-locus to a low recombination region on chromosome I. 

S-RNase KO lines were obtained using a dual sgRNA strategy in which premature stop codons 

were generated. After self-pollination, fruits were set in selected KO lines in T0 and T1 lines. 

Cas9-free KO lines were also identified in T1 lines. Our results demonstrated the inheritance and 

stablity of the S-RNase KO phenotype, which can contribute to utilization of SC as a first step for 

the generation of commercial diploid cultivars. 

 

 

 

109 

 

APPENDIX

 

110 

 

Figure S3.1. S-RNase open reading-frames in T0 knock-out (KO) lines derived in two SI diploid potato lines. Sequences shown within 

the dotted lines contain 5’ and 3’ mRNA borders of this gene. Primer sequences designed to amplify the S-RNase ORF are shown as 

well as the sgRNA 1 located in exon 1. Individual dots represent nucleotides within the S-RNase gene. All T0 KO lines had a 

frameshift near the sgRNA target region (blue dotted line) creating a premature stop codon (red dotted line). Black dotted line 

represents in frame S-RNase regions. nucleotides. Black dotted dash lines represent missing transcript sequence. S-RNase alleles from 

 

DM and RH are shown for each KO line. 

 

 

111 

 

Table S3.1. Retrieved Illumina reads for genomic assembly of the DRH potato lines and 

assembly statistics. 

Tissue 

DRH-195 

DRH-310 

Mature leaflet  SRR4018191 
SRR4018147 
Leaf 
Tuber 
SRR4018170 
                      Total: 

SRA accession  Pair-end Reads  SRA accession  Pair-end Reads 
24,891,913 
29,113,282 
31,283,414 
85,288,609 

30,799,234  SRR4018197 
30,423,428  SRR4018153 
31,800,294  SRR4018174 
93,022,956 

  

 
 

 

 

112 

Table S3.2. Primers and sgRNAs used in this study. 

Sequence 

ATGTTTAAATCACTGCTTACATCAAC 
TCAGGGACGGAAAAATATTTTCCCTG 

Description 
S-RNase forward primer 
S-RNase reverse primer 

Name 
S-RNase F 
S-RNase R 
S-RNaseRH F  GTTTTGTTTAATTTACTGAAAAGCTTA  RH-specific S-RNase forward primer 
S-RNaseRH R  AAAGATTTCTTCAAATGTATAACT 
RH-specific S-RNase reverse primer 
Elongation factor one alpha forward 
EF1a F 
GGTGGTTTTGAAGCTGGTATCTCT 
primer 
Elongation factor one alpha reverse 
primer 
Single-guide RNA 1 targeting exon 1 
Single-guide RNA 2 targeting exon 2 

CCAGTAGGGCCAAAGGTCACA 

EF1a R 
sgRNA1 
sgRNA2 
* In bold, protospacer adjacent motif (PAM) sequence 
 
 

AATTGCAACTGGTATTAACATGG* 
CCTGATATCAAGTGTACTGAAGG 

 

 

113 

Table S3.3. Top TBLASTN hits of reported S-RNase proteins in the DM genome assembly.  

Species 

Scaffold 

HSP* 
Number 

Hit 
Score 

E value  Length  Percent 

Id 

Nicotiana sylvestris 

(CAA05306.1) 

Petunia integrifolia  

(AAG21384.1) 
Petunia x hybrida  
(AAA33729.1) 

Solanum chacoense 

(AAA50306.1) 
Solanum chilense 
(BAC00934.1) 
Solanum neorickii 
(BAC00940.1) 

Solanum tuberosum 

(Q01796.1) 

chr00 

chr00 

chr00 

chr00 

chr00 

chr00 

chr00 

* High-scoring Segment Pair 
 

1 

1 

1 

1 

1 

1 

1 

192 

159 

226 

415 

313 

321 

204 

8.00E-55 

244 

41.8 

3.00E-43 

222 

44.14 

2.00E-66 

251 

48.21 

3.00E-
132 

245 

84.08 

1.00E-96 

244 

63.93 

2.00E-99 

244 

64.75 

1.00E-58 

230 

49.57 

 

 

 

 

 

 

 

 

 

 

114 

Table S3.4. Fruit set and seed count upon self-pollination in DRH-195 wild type and S-RNase-

Self-

pollinated 
flowers** 

Total fruit 

set 

Ratio of 

fruit set per 
pollination 

Number of 
cluster and 
seeds per 
cluster 

130 

4 

29 

0.03 

0.76 

derived KO lines. 

Line 

DRH-195 
(WT)* 

DRH-195.105 

38 

DRH-195.128 

80 

9 

0.11 

DRH-195.137 

29 

14 

0.48 

DRH-195.142 

125 

36 

0.29 

DRH-195.158 

121 

46 

0.38 

 

115 

50 

200 
60 
240 
19 
7 
21 
6 
10 
11 
80 
24 
34 
50 
20 
30 
40 
50 
250 
30 
50 
100 
40 
50 
38 
19 
51 
14 
38 
45 

Table S3.4. (cont’d) 
 

 

 

 

 

DRH-195.160 

71 

23 

0.32 

*Wild type. ** Total of flowers in two replicates. 

39 
45 
226 
50 
50 
200 
29 
8 
33 
11 
15 
200 

 

 

 

116 

Table S3.5. Fruit set and seed count upon self-pollination in DRH-310 wild type and S-RNase-

derived KO lines.  

Line 

Self-

pollinated 
flowers** 

Total fruit 

set 

Ratio of 

fruit set per 
pollination 

DRH-310 
(WT)* 

DRH-310.8 

123 

38 

0 

5 

0 

0.13 

DRH-310.21 

52 

16 

0.31 

DRH-310.33 

53 

15 

0.28 

* Wild type. ** Total of flowers in two replicates 

 

 Number of 
cluster and 
seeds per 
cluster 

0 

2 
3 
15 
50 
50 
20 
28 
50 
63 
10 
19 
20 
20 
18 
10 

 
 
 
 
 
 
 
 
 
 
 
 
 

 

117 

LITERATURE CITED 

 

118 

LITERATURE CITED 

 
 
 

Altschul S. F., Gish W., Miller W., Myers E. W., Lipman D. J. (1990). Basic Local Alignment 

Search Tool. J. Mol. Biol. 215: 403–410. 

Bortesi L., Fischer R. (2015). The CRISPR/Cas9 system for plant genome editing and beyond. 

Biotechnol. Adv. 33: 41–52. 

Brisolara-Corrêa L., Thompson C. E., Fernandes C. L., Freitas L. B. de. (2015). Diversification 

and distinctive structural features of S-RNase alleles in the genus Solanum. Mol. Genet. 
Genomics. 290: 987–1002. 

Bush W. S., Moore J. H. (2012). Chapter 11: Genome-Wide Association Studies (F Lewitter 

and M Kann, Eds.). PLoS Computational Biology 8: e1002822. 

Canto J. Do, Studer B., Lubberstedt T. (2016). Overcoming self-incompatibility in grasses: a 

pathway to hybrid breeding. Theor. Appl. Genet. 129: 1815–1829. 

Covey P. A., Kondo K., Welch L., Frank E., Sianta S., Kumar A., Nuñez R., Lopez-Casado G., 
Knaap E. Van Der, Rose J. K. C., McClure B. A., Bedinger P. A. (2010). Multiple features 
that distinguish unilateral incongruity and self-incompatibility in the tomato clade. The 
Plant Journal 64: 367–378. 

Devaux A., Kromann P., Ortiz O. (2014). Potatoes for sustainable global food security. Potato 

Research 57: 185–199. 

Doudna J. A., Charpentier E. (2014). The new frontier of genome engineering with CRISPR-

Cas9. Science 346: 1258096. 

Dzidzienyo D. K., Bryan G. J., Wilde G., Robbins T. P. (2016). Allelic diversity of S-RNase 

alleles in diploid potato species. Theor. Appl. Genet. 129: 1985–2001. 

FAOSTAT. (2019). FAOSTAT database collections. Food and Agriculture Organization of the 

United Nations. 

Feng Z., Mao Y., Xu N., Zhang B., Wei P., Yang D.-L., Wang Z., Zhang Z., Zheng R., Yang 

L., Zeng L., Liu X., Zhu J.-K. (2014). Multigeneration analysis reveals the inheritance, 
specificity, and patterns of CRISPR/Cas-induced gene modifications in Arabidopsis. Proc. 
Natl. Acad. Sci. U S A. 111: 4632–4637. 

Fujii S., Kubo K. I., Takayama S. (2016). Non-self- and self-recognition models in plant self-

incompatibility. Nat. Plants 2: 16130. 

 

119 

Gebhardt C., Ritter E., Barone A., Debener T., Walkemeier B., Schachtschabel U., Kaufmann 
H., Thompson R. D., Bonierbale M. W., Ganal M. W., Tanksley S. D., Salamini F. (1991). 
RFLP maps of potato and their alignment with the homoeologous tomato genome. Theor. 
Appl. Genet. 83: 49–57. 

Goldraij A., Kondo K., Lee C. B., Hancock C. N., Sivaguru M., Vazquez-Santana S., Kim S., 
Phillips T. E., Cruz-Garcia F., McClure B. (2006). Compartmentalization of S-RNase and 
HT-B degradation in self-incompatible Nicotiana. Nature 439: 805. 

Gomez M. A., Lin Z. D., Moll T., Chauhan R. D., Hayden L., Renninger K., Beyene G., Taylor 

N. J., Carrington J. C., Staskawicz B. J., Bart R. S. (2018). Simultaneous CRISPR/Cas9-
mediated editing of cassava eIF4E isoforms nCBP-1 and nCBP-2 reduces cassava brown 
streak disease symptom severity and incidence. Plant. Biotechnol. J.: 1–14. 

Hadj-Arab H., Chèvre A.-M., Gaude T., Chable V. (2010). Variability of the self-

incompatibility reaction in Brassica oleracea L. with S15 haplotype. Sex. Plant. Reprod. 
23: 141–151. 

Hall T. A. (1999). BioEdit: a user-friendly biological sequence alignment editor and analysis 

program for Windows 95/98/NT. Nucleic Acids Symposium Series 41: 95–98. 

Hardigan M. A., Crisovan E., Hamiltion J. P., Kim J., Laimbeer P., Leisner C. P., Manrique-

Carpintero N. C., Newton L., Pham G. M., Vaillancourt B., Yang X., Zeng Z., Douches D., 
Jiang J., Veilleux R. E., Buell C. R. (2016). Genome reduction uncovers a large dispensable 
genome and adaptive role for copy number variation in asexually propagated Solanum 
tuberosum. Plant Cell 28: 388–405. 

Hosaka K., Hanneman R. E. (1998). Genetics of self-compatibility in a self-incompatible wild 
diploid potato species Solanum chacoense. 1. Detection of an S locus inhibitor (Sli) gene. 
Euphytica 99: 191–197. 

Hua Z.-H., Fields A., Kao T. (2008). Biochemical Models for S-RNase-Based Self-

Incompatibility. Molecular Plant 1: 575–585. 

Igic B., Kohn J. R. (2001). Evolutionary relationships among self-incompatibility RNases. Proc. 

Natl. Acad. Sci. U S A. 98: 13167–13171. 

Ioerger T. R., Clark A. G., Kao T. H. (1990). Polymorphism at the self-incompatibility locus in 

Solanaceae predates speciation. Proc. Natl. Acad. Sci U S A. 87: 9732 LP – 9735. 

Ioerger T. R., Gohlke J. R., Xu B., Kao T. I. H. (1991). Sexual Plant lleproduction Primary 

structural features of the self-incompatibility protein in Solanaceae. Sex. Plant. Reprod 4: 
81–87. 

Jaganathan D., Ramasamy K., Sellamuthu G., Jayabalan S., Venkataraman G. (2018). CRISPR 

for crop improvement: An update review. Front. Plant. Sci. 9: 985. 

 

120 

Jansky S. H., Chung Y. S., Kittipadukal P. (2014). M6: A diploid potato inbred line for use in 

breeding and genetics research. J. Plant. Regist. 8: 195–199. 

Jansky S. H., Charkowski A. O., Douches D. S., Gusmini G., Richael C., Bethke P. C., Spooner 

D. M., Novy R. G., Jong H. De, Jong W. S. De, Bamberg J. B., Thompson A. L., 
Bizimungu B., Holm D. G., Brown C. R., Haynes K. G., Sathuvalli V. R., Veilleux R. E., 
Miller J. C., Bradeen J. M., Jiang J. (2016). Reinventing potato as a diploid inbred line–
based crop. Crop Science 56: 1412–1422. 

Kao T., Tsukamoto T. (2004). The molecular and genetic bases of S-RNase-based self-

incompatibility. Plant Cell 16: S72–S83. 

Karp A., Risiott R., Jones M. G. K., Bright S. W. J. (1984). Chromosome doubling in 

monohaploid and dihaploid potatoes by regeneration from cultured leaf explants. Plant Cell 
Tissue Organ Cult. 3: 363–373. 

Koncz C., Martini N., Szabados L., Hrouda M., Bachmair A., Schell J., 1994 Specialized 

vectors for gene tagging and expression studies. In: Gelvin SB, Schilperoort RA (Eds.), 
Plant Molecular Biology Manual, Springer Netherlands, Dordrecht, pp. 53–74. 

Kowyama Y., Kunz C., Lewis I., Newbigin E., Clarke A. E., Anderson M. A. (1994). Self-

compatibility in a Lycopersicon peruvianum variant (LA2157) is associated with a lack of 
style S-RNase activity. Theoretical and Applied Genetics 88: 859–864. 

Kubo K., Entani T., Takara A., Wang N., Fields A. M., Hua Z., Toyoda M., Kawashima S., 

Ando T., Isogai A., Kao T., Takayama S. (2010). Collaborative non-self recognition system 
in S-RNase based self-incompatibility. Science 330: 796–799. 

Kubo K., Paape T., Hatakeyama M., Entani T., Takara A., Kajihara K., Tsukahara M., Shimizu-

Inatsugi R., Shimizu K. K., Takayama S. (2015). Gene duplication and genetic exchange 
drive the evolution of S-RNase-based self-incompatibility in Petunia. Nat. Plants 1: 14005. 

Kumar S., Stecher G., Tamura K. (2016). MEGA7: Molecular evolutionary genetics analysis 

version 7.0 for bigger datasets. Mol. Biol. Evol. 33: 1870–1874. 

Li W., Zarka K. A., Douches D. S., Coombs J. J., Pett W. L., Grafius E. J. (1999). Coexpression 

of potato PVYo coat protein and cryV-Bt genes in potato. J. Amer. Soc. Hort. Sci. 124: 
218–223. 

Li H., Handsaker B., Wysoker A., Fennell T., Ruan J., Homer N., Marth G., Abecasis G., 

Durbin R., Subgroup 1000 genome project data processing. (2009). The sequence 
alignment/map format and SAMtools. Bioinformatics 25: 2078–2079. 

Li H. (2011). A statistical framework for SNP calling, mutation discovery, association mapping 

and population genetical parameter estimation from sequencing data. Bioinformatics 27: 
2987–2993. 

 

121 

Li H. (2013). Aligning sequence reads, clone sequences and assembly contigs with BWA-

MEM. eprint arXiv:1303.3997: arXiv:1303.3997. 

Li W., Chetelat R. T. (2015). Unilateral incompatibility gene ui1.1 encodes an S-locus F-box 

protein expressed in pollen of Solanum species. Proc. Natl. Acad. Sci. U S A. 112: 4417–22. 

Liao J., Dai J., Kang H., Liao K., Ma W., Wang J., Chen S. (2016). Plasticity in the self-

incompatibility system of cultivated Nicotiana alata. Euphytica 208: 129–141. 

Manrique-Carpintero N. C., Coombs J. J., Cui Y., Veilleux R. E., Buell C. R., Douches D. 

(2015). Genetic map and QTL analysis of agronomic traits in a diploid potato population 
using single nucleotide polymorphism markers. Crop Science 55: 2566–2579. 

Matton D. P., Maes O., Laublin G., Xike Q., Bertrand C., Morse D., Cappadocia M. (1997). 

Hypervariable domains of self-incompatibility RNases mediate allele-specific pollen 
recognition. Plant Cell 9: 1757–1766. 

Matton D. P., Luu D. T., Xike Q., Laublin G., O’Brien M., Maes O., Morse D., Cappadocia M. 
(1999). Production of an S RNase with dual specificity suggests a novel hypothesis for the 
generation of new S alleles. Plant Cell 11: 2087–2097. 

McClure B., Mou B., Canevascini S., Bernatzky R. (1999). A small asparagine-rich protein 

required for S-allele-specific pollen rejection in Nicotiana. Proc. Natl. Acad. Sci. U S A. 96: 
13548–13553. 

Mena-Ali J. I., Stephenson A. G. (2007). Segregation analyses of partial self-incompatibility in 

self and cross progeny of Solanum carolinense reveal a leaky S-allele. Genetics 177: 501–
510. 

Mendoza H. A., Mihovilovich E. J., Saguma F. (1996). Identification of triplex (YYYy) Potato 
Virus Y (PVY) immune progenitors derived from Solanum tuberosum ssp. andigena. Am. 
Potato J. 73: 13–19. 

Miller J. S., Kostyun J. L. (2011). Functional gametophytic self-incompatibility in a peripheral 

population of Solanum peruvianum (Solanaceae). Heredity 107: 30–39. 

Murashige T., Skoog F. (1962). A revised medium for rapid growth and bio assays with 

tobacco tissue cultures. Physiol. Plant 15: 473–497. 

O’Brien M., Kapfer C., Major G., Laurin M., Bertrand C., Kondo K., Kowyama Y., Matton D. 
P. (2002). Molecular analysis of the stylar-expressed Solanum chacoense small asparagine-
rich protein family related to the HT modifier of gametophytic self-incompatibility in 
Nicotiana. Plant J 32: 985–996. 

Ordoñez B. (2014a). Pollen Viability Assessment. International Potato Center (CIP). 8p, Lima, 

Peru. 

 

122 

Ordoñez B. (2014b). Assessment of ploidy by chloroplast count in stomatal guard cells. : 4p. 

Owen H. R., Veilleux R. E., Levy D., Ochs D. L. (1988). Environmental, genotypic, and ploidy 
effects on endopolyploidization within a genotype of Solanum phureja and its derivatives. 
Genome 30: 506–510. 

Park J., Bae S., Kim J. (2015). Cas-Designer: a web-based tool for choice of CRISPR-Cas9 

target sites. Bioinformatics 31: 4014–4016. 

Pellagatti A., Dolatshad H., Valletta S., Boultwood J. (2015). Application of CRISPR/Cas9 

genome editing to the study and treatment of disease. Arch. Toxicol. 89: 1023–1034. 

Peterson B. A., Holt S. H., Laimbeer F. P. E., Doulis A. G., Coombs J., Douches D. S., 

Hardigan M. A., Buell C. R., Veilleux R. E. (2016). Self-fertility in a cultivated diploid 
potato population examined with the Infinium 8303 Potato Single-Nucleotide 
Polymorphism Array. The Plant Genome 9: 1–13. 

Porcher E., Lande R. (2005). Loss of gametophytic self-incompatibility with evolution of 

inbreeding depression. Evolution 59: 46–60. 

Ramanauskas K., Igić B. (2017). The evolutionary history of plant T2/S-type ribonucleases. 

PeerJ 5: e3790. 

Raymundo R., Asseng S., Robertson R., Petsakos A., Hoogenboom G., Quiroz R., Hareau G., 

Wolf J. (2017). Climate change impact on global potato production. Eur. J. Agron.: 
j.eja.2017.11.008. 

Royo J., Kunz C., Kowyama Y., Anderson M., Clarke A. E., Newbigin E. (1994). Loss of a 

histidine residue at the active site of S-locus ribonuclease is associated with self-
compatibility in Lycopersicon peruvianum. Proc Natl Acad Sci U S A 91: 6511–6514. 

Saba-El-Leil M. K., Rivard S., Morse D., Cappadocia M. (1994). The S11 and S13 self 

incompatibility alleles in Solanum chacoense Bitt. are remarkably similar. Plant. Mol. Biol. 
24: 571–583. 

Sassa H., Kakui H., Miyamoto M., Suzuki Y., Hanada T., Ushijima K., Kusaba M., Hirano H., 

Koba T. (2007). S locus F-box brothers: multiple and pollen-specific F-box genes with S 
haplotype-specific polymorphisms in apple and Japanese pear. Genetics 175: 1869–1881. 

Sievers F., Wilm A., Dineen D., Gibson T. J., Karplus K., Li W., Lopez R., McWilliam H., 

Remmert M., Söding J., Thompson J. D., Higgins D. G. (2011). Fast, scalable generation of 
high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7: 
539. 

 

123 

Sijacic P., Wang X., Skirpan A. L., Wang Y., Dowd P. E., McCubbin A. G., Huang S., Kao T. 

H. (2004). Identification of the pollen determinant of S-RNase-mediated self-
incompatibility. Nature 429: 302–305. 

Soulard J., Boivin N., Morse D., Cappadocia M., 2014 eEF1A is an S-RNase binding factor in 

self-incompatible Solanum chacoense. PloS One 9: e90206. 

Stephenson A. G., Good S. V, Vogler D. W. (2000). Interrelationships among inbreeding 

depression, plasticity in the self-incompatibility system, and the breeding system of 
Campanula rapunculoides L. (Campanulaceae). Ann. Bot. 85: 211–219. 

Sternberg S. H., Redding S., Jinek M., Greene E. C., Doudna J. A. (2014). DNA interrogation 

by the CRISPR RNA-guided endonuclease Cas9. Nature 507: 62–7. 

Stone J. L., Sasuclark M. A., Blomberg C. P. (2006). Variation in the self-incompatibility 

response within and among populations of the tropical shrub Witheringia solanacea 
(Solanaceae). Am. J. Bot. 93: 592–598. 

Takayama S., Isogai A. (2005). Self-Incompatibility in Plants. Annu. Rev. Plant. Biol. 56: 467–

489. 

The Potato Genome Sequencing Consortium. (2011). Genome sequence and analysis of the 

tuber crop potato. Nature 475: 189–195. 

Wang W., Pan Q., He F., Akhunova A., Chao S., Trick H., Akhunov E. (2018). 

Transgenerational CRISPR-Cas9 activity facilitates multiplex gene editing in allopolyploid 
wheat. The CRISPR Journal 1: 65–74. 

Webb M. C., Williams E. G. (1988). Effects of temperature, light, nutrients and carbon dioxide 

on the strength of the self-incompatibility response in detached flowers of Lycopersicon 
peruvianum. Ann. Bot. 61: 395–404. 

Xing H.-L., Dong L., Wang Z.-P., Zhang H.-Y., Han C.-Y., Liu B., Wang X.-C., Chen Q.-J. 
(2014). A CRISPR/Cas9 toolkit for multiplex genome editing in plants. BMC Plant. Biol. 
14: 327. 

Xu R.-F., Li H., Qin R.-Y., Li J., Qiu C.-H., Yang Y.-C., Ma H., Li L., Wei P.-C., Yang J.-B. 
(2015). Generation of inheritable and “transgene clean” targeted genome-modified rice in 
later generations using the CRISPR/Cas9 system. Sci. Rep. 5: 11491. 

Ye M., Peng Z., Tang D., Yang Z., Li D., Xu Y., Zhang C., Huang S. (2018). Generation of 

self-compatible diploid potato by knockout of S-RNase. Nat. Plants 4: 651–654. 

 

 

 

 

124 

Chapter 4  

VALIDATION OF THE RB-MEDIATED RESISTANCE SUPPRESSION BY THE IPI-

O4 EFFECTOR FROM P. infestans IN POTATO 

 

125 

 

 

 

I. Abstract 

Potato (Solanum tuberosum L.) is considered the third most important food crop after rice and 

wheat. However, pathogens such as late blight (Phytophthora infestans (Mont.) de Bary), can 

generate annual losses over 16% of the total global potato production. The most effective way to 

control this pathogen is through the introgression of resistant genes from landraces or wild-

relative species. In particular, resistance to the late blight gene (RB) from S. bulbocastanum, with 

a Coiled-Coil (CC)-Nucleotide-binding Leucine-rich Repeat (NLR) gene architecture, confers a 

wide-spectrum resistance to late blight in transgenic potato lines. Under pathogen attack, the late 

blight IPI-O1 effector elicits the RB dimerization through the CC domain, triggering a disease 

resistance response, while IPI-O4 suppress the IPI-O1-mediated resistance to late blight. In this 

study, we used in vivo and in vitro approaches to confirm the molecular interactions leading to 

the activation or suppression of late blight resistance responses through CC/IPI-O protein 

interactions. We observed a hypersensitive response (HR) when RB was co-infiltrated with IPI-

O1 via N. benthamiana agroinfiltration. However, no HR was observed when IPI-O1 or IPI-O4 

were co-infiltrated using a synthetic RB gene carrying a CC domain from S. pinnatisectum (RB-

SP). In contrast, HR was elicited when IPI-O1 and IPI-O4 were infiltrated in a RB-SP transgenic 

line. Moreover, this study was unable to demonstrate CC-self association using a yeast-two-

hybrid system. These results suggest that in addition to RB, other proteins or RB domains may be 

involved in resistance response in vivo. Further analysis should be conducted to confirm non 

self-CC dimerization and its relationship with observed HR mediated by unreported CC/IPI-O 

interactions. 

 

 

 

126 

II. Introduction  

Cultivated potato (Solanum tuberosum L.) is the third most important food crop after rice and 

wheat (Devaux et al., 2014) and plays an essential role in human nutrition as a primary source of 

carbohydrates. This staple food is also known for its high content of phenolic compounds, 

opening a new window for the potato’s commercialization as a functional food, due to its 

potential anti-oxidative activity (Friedman, 1997; Vinson et al., 2012). Although global 

production of potato totaled 388 million tons in 2017 (FAOSTAT, 2019), potatoes face major 

production losses due to biotic and abiotic stresses that will increase with global warming 

(Raymundo et al., 2017). 

In particular, pathogens such as late blight (Phytophthora infestans (Mont.) de bary), 

represent one of the most devastating diseases for this crop, infecting vegetative tissues and 

killing the entire plant within 7 to 10 days after infection. Late blight accounts for annual losses 

of 16% of total global potato production (Haverkort et al., 2009). Additionally, under the current 

fluctuating weather conditions, it is expected that late blight incidences will increase worldwide, 

affecting mainly highlands in developing countries in upcoming decades (Sparks et al., 2014). 

Late blight is a hemibiotrophic oomycete that infects vegetative plant tissues through 

multinucleate sporangium. The asexual stage uses mononucleate zoospores for germination, 

produced from cleaved sporangia, entering into the host via stomata, wounds or appressorium 

formation (Leesutthiphonchai et al., 2018). However, sexual spores, known as oospores, are also 

produced and requires two mating types; A1 and A2. This system increases late blight diversity 

through sexual recombination when both mating types are present in a population, favoring its 

adaptation to extreme environmental conditions (Smart and Fry, 2001). 

 

 

127 

Although pathogen infection in late blight of potato can be controlled using protectants or 

systemic fungicides, they are ineffective if environmental conditions favor pathogen dispersion 

(Nowicki et al., 2011) and also promote the emergence of fungicide-resistant strains (Pomerantz 

et al., 2014). The most effective way to control the incidence of late blight in potatoes is through 

the generation of resistant varieties (Ahn and Park, 2013). However, this process can take up to 

15 years, which is time-consuming and ineffective against fast-evolving pathogens such as late 

blight (Lozano et al., 2012; The Potato Genome Sequencing Consortium, 2011). 

Due to their close phylogenetic relationship, wild tuber bearing-species have been the 

primary source of resistance (R) genes for late blight in cultivated potato (Rodewald and 

Trognitz, 2013). Multiple R genes have been discovered in several potato wild relatives. In 

particular, the late blight resistance gene (RB) from S. bulbocastanum (blb), confers broad-

spectrum disease resistance against this pathogen (Song et al., 2003). Although partial resistance 

in has been observed field, RB-transgenic potatoes present an increase in foliar resistance in 

growth chamber and greenhouse experiments when compared with wild type genotypes, without 

significant effects on tuber yield (Halterman et al., 2008). 

The RB gene encodes a modular resistance protein comprising an N-terminal Coiled-Coil 

(CC) domain and a Nucleotide-binding Leucine-rich Repeat (NLR) domain core (Song et al., 

2003). RB remains in a resting state prior to late blight infection. After pathogen attack, a 

conformational change elicited by the recognition of the IPI-O1 effector from the late blight 

pathogen, results in RB dimerization through the CC domain, leading to a hypersensitive 

response (HR) (Chen et al., 2012). However, when the IPI-O4 effector is present, RB-mediated 

resistance is suppressed by direct competition with IPI-O1, preventing the interaction of the CC 

domains or other resistance-related signaling components (Chen et al., 2012; Chen and 

 

128 

Halterman, 2017). 

The CC domain from the Solanum pinnatisectum (pnt) homologous RB gene does not exhibit 

any interaction with either IPI-O1 or IPI-O4, suggesting that this species may escape IP-O4-

mediated resistance suppression (Chen et al., 2012). In an attempt to generate a durable late 

blight resistance response in RB-transgenic potato, we validated the IPI-O4 mediated suppression 

in vitro and in vivo. Agrobacterium-infiltration (agroinfiltration) assays in Nicotiana 

benthamiana using a synthetic RB gene (RB-SP) carrying a CC domain from S. pinnatisectum 

that led to a lack of HR when challenged with IPI-O effectors, confirming previously reported 

results. However, HR was observed when IPI-O1 and IPI-O4 were infiltrated in a RB-SP 

transgenic potato line, suggesting that additional mechanisms may be involved in a resistance 

response in vivo. Moreover, CC dimerization was not observed in yeast-two-hybrid (Y2H) 

assays. Further work is required to confirm protein expression in yeast and validate the role of 

the IPI-O effectors on the activation/suppression of late blight resistance in potatoes. 

 

III. Material and Methods 

i. 

Plant material 

The diploid potato DRH S5 28-5 line (referred hereafter as DRH-S5) was used in this study. This 

line was generated from five rounds of self-pollination using a F1 founder from a cross between 

the doubled monoploid, S. tuberosum Group Phureja DM1-3 516 R44, and a heterozygous 

diploid breeding line, S. tuberosum Group Tuberosum RH89-039-16. Plants were maintained in 

vitro, propagated on Murashige and Skoog (MS), medium (MS basal salts plus vitamins, 3% 

sucrose, 0.7% plant agar, pH 5.8) (Murashige and Skoog, 1962) and cultured in a growth 

 

129 

chamber with 16-h-light/8-h-night photoperiod at 22°C and average light intensity of 200 μmoles 

m-2s-1. 

 

ii. 

RB and IPI-O constructs and Agrobacterium-mediated transformation 

A plasmid containing a synthetic RB gene with a CC domain from pnt and a NLR domain core 

from blb (referred here after as RB-SP) along with Agrobacterium tumefaciens cultures 

containing the late blight IPI-O1 and IPI-O4 effectors (Chen et al., 2012) were kindly donated by 

Dr. Halterman (University of Wisconsin), and used in this study. The pSPUD69 construct, 

carrying the RB gene from S. bulbocastanum under control of its native promoter and terminator 

(Kuhl et al., 2007), was used as a control. RB-SP and RB constructs with the NPTII gene as a 

selectable marker, were transferred into A. tumefaciens strain GV3101 pMP90 (Koncz et al., 

1994) by electroporation. 

Agrobacterium-mediated transformations were performed using leaf segments from four-

week-old tissue culture plants of DRH-S5 as described by Li et al. (1999). Briefly, explants were 

pre-cultured on a step I media (MS salts, 3% sucrose, 5 g/l phytoagar, 1 mg/l thiamine-HCl, 0. 8 

mg/l zeatin-riboside and 2 mg/l 2,4-D) for four days and inoculated with A. tumefaciens carrying 

the RB genes. After three days, explants were rinsed with sterile distilled water containing 250 

mg/l cefotaxime and 200 mg/l carbenicillin and placed onto step II media (MS salts, 3% sucrose, 

5 g/L phytoagar, 1 mg/L thiamine-HCl, 0.8 mg/L zeatin-riboside, 2 mg/L gibberellic acid, 50 

mg/L kanamycin and 150 mg/L ticarcillin disodium and clavulanate potassium). Explants were 

transferred to fresh step II media every week. After approximately 30 days, transformation 

events were selected from step II media and transferred to root induction media containing MS 

medium supplemented with antibiotics for selection as described above. 

 

130 

iii. 

Leaf Agroinfiltration on N. benthamiana and transgenic potatoes 

Four week-old N. benthamiana leaves were co-infiltrated with the IPI-O effectors and RB-SP/RB 

genes as described by Ma et al. (2012). Briefly, A. tumefaciens containing the RB/RB-SP and the 

IPI-O1/IPI-O4 constructs were inoculated in LBmani media (10 g/L Bacto-tryptone, 5 g/L yeast 

extract, 2.5 g/L sodium chloride, 10 g/L mannitol) supplemented with 20 mM acetosyringone, 10 

mM 2-(n-morpholino)ethanesulfonic acid (MES) and 50 mg/L kanamycin. Cells were harvested 

when cultures reached an optical density at 600 nm (OD600) of 0.8 and suspended in MMAi 

medium (5 g/L MS salts, 20 g/L sucrose, 200 µM acetosyringone and 10 mM MES). For HR 

induction, Agrobacterium suspensions containing either RB-SP or RB were syringe-infiltrated at 

the abaxial side of each leaf. After drying, suspensions containing IPI-O effectors were syringe-

infiltrated overlapping the RB-SP and RB infiltrations respectively, for a total of four R/Avr gene 

interactions: RB/IPI-O1, RB/IPI-O4, RB-SP/IPI-O1, and RB-SP/IPI-O4. The RB/IPI-O1 co-

infiltration and an empty A. tumefaciens inoculum were used as a positive and negative control, 

respectively. 

One month-old in-vitro DRH transgenic plants were planted in 10 cm square plastic pots with 

a peat and perlite growth medium mixture (Bacto professional planting mix) and placed into a 

grow chamber with a light intensity of 250 µmoles m−2 s−1, 16/8-h light/dark photoperiod and a 

temperature of 25 °C. Plants were fertilized with Peters Professional® 20: 20: 20 fertilizer (The 

Scotts Co., Marysville, OH) at a rate of 500 mg/l twice a week. Three weeks later, plants were 

agroinfiltrated as described above with the IPI-O4 effector using the IPI-O1 and an empty A. 

tumefaciens inoculum as a positive and negative control, respectively. For both assays, at least 

three independent leaves were agroinfiltrated. Plants were kept under growth chamber conditions 

and the HR response was assessed one week after infiltration. 

 

131 

iv. 

Molecular characterization of RB-SP and RB transgenic lines. 

DNA was isolated from young leaves using the DNeasy Plant Mini Kit (Qiagen, Hilden, 

Germany). NPTII and RB-derived primers (Colton et al., 2006) were used to select positive 

transformation events (Table 4.1). PCR was carried out using GoTaq DNA polymerase 

(Promega, Fitchburg, WI) with the following thermocycler conditions: one cycle of initial 

denaturation for 4 min at 94 °C, followed by 34 cycles for 15 s at 94 °C, 45 s at 55 °C and 1 min 

at 72 °C and a final extension of 5 min at 72 °C. Amplicons were visualized on 1% (w/v) agarose 

gels. 

 

Table 4.1. Primers used in this study 

Name 
RB-F 
RB-R 
NPTII-F 
NPTII-R 
SUS3-F 
SUS3-R 

 

Description 
Sequence 
RB forward primer 
CACGAGTGCCCTTTTCTGAC 
RB reverse primer 
ACAATTGAATTTTTAGACTT 
ATGATTGAACAAGATGGATTGCAC 
NPTII (kanamycin) forward primer 
NPTII (kanamycin) reverse primer 
CCAAGCTCTTCAGCAATATCACGG 
Sucrose synthase gene forward primer 
CTGCAAGCTAAGCCTGATCTTATTAT 
TTCGGAGTATGGAAAATAGAGATTCA  Sucrose synthase gene reverse primer 

Expression analyses were performed on transformation events carrying either the RB-SP or RB 

transgenes. Total RNA was isolated from six-week-old plant leaves using the RNeasy Plant Mini 

Kit (Qiagen, Hilden, Germany) and DNase treated using the TURBO DNA-free kit 

(ThermoFisher, Carlsbad, CA) following the manufacturer’s instructions. RNA was quantified 

using a NanoDrop spectrophotometer (ThermoFisher Scientific, Grand Island, NY) and reverse-

transcription polymerase chain reaction (RT-PCR) was carried-out using 1 µg of total RNA 

using the Super-Script One-Step RT-PCR kit (Thermo Fisher Scientific, Carlsbad, CA). The RB 

primers, designed to amplify a 218 bp of the NLS core domain, and the Sucrose Synthase 3 Gene 

 

132 

(SUS3) housekeeping internal control, were used for the RT-PCR reaction (Table 4.1). DRH-S5 

(WT) were used as a negative control. 

 

v. 

Protein interaction assays  

The Matchmaker® Gold Y2H System (Takara Bio USA) was used for protein interaction 

screening. Full-length coding regions of the blb and pnt CC domains (495 bp each) were 

individually cloned into the multiple cloning site (MCS) of the pBridge and pGADT7 vectors, 

encoding a bait (Gal4-BD) and prey (Gal4-AD) fusion proteins, respectively. Likewise, the 

coding region of the IPI-O4 effector (396 bp) was cloned into a second MCS within the pBridge 

vector. Both vectors were co-transformed using a lithium acetate procedure into the Y2HGold 

haploid yeast strain according to the Matchmaker user manual. Co-transformants were selected 

by culturing in double dropout minimum (SD) medium (DDO) lacking leucine (-Leu) and 

tryptophan (-Trp) at 30°C for 4-5 days. Selected yeast colonies were sub-cultured in DDO 

supplemented with X-a-Gal and Aureobasidin A (DDO/X/A). Healthy 2 mm blue colonies were 

patched onto the higher stringency quadruple dropout minimum medium (QDO) lacking adenine 

(-Ade), Histidine (-His), leucine (-Leu) and tryptophan (-Trp), supplemented with X-a-Gal and 

Aureobasidin A (QDO/X/A). For IPI-O4 expression, healthy 2 mm colonies were patched from 

DDO onto DDO/X/A and QDO/X/A media lacking methionine (-Met). The pGBKT7-

53/pGAT7-T and pGBKT7-53/ pGBKT7-Lam vectors were co-transformed and used as positive 

and negative controls, respectively. Autoactivation tests were conducted using an empty pBridge 

vector co-transformed with pGADT7+CC. 

 

 

 

133 

IV. Results 

i. 

HR-induced assessment in N. benthamiana 

N. benthamiana leaves were co-infiltrated to test whether the synthetic RB gene interacts with 

IPI-O4 in vivo. Co-infiltrated areas with RB and IPI-O1 triggered HR as expected. However, no 

HR was observed when RB-SP and IPI-O4 were co-infiltrated along with RB-SP/IPI-O1, RB/IPI-

O4 and empty control vector (Figure 4.1). These results are consistent with those of Chen et al. 

(2012) who reported a lack of interaction between the CC domain from pnt and the IPI-O 

effectors using in vitro approaches. These findings demonstrate that RB-SP is not able to 

recognize either IPI-O1 or IPI-O4 effectors via the CC domain when co-expressing in a 

heterologous system. 

Figure 4.1. Induction of hypersensitive response (HR) in Nicotiana benthamiana. Agrobacterium 

 

tumefaciens expressing RB, RB-PS, IPI-O1 and IPI-O4 were co-infiltrated into N. benthamiana 

leaves. HR was observed when RB was co-infiltrated with IPI-O1 (used as a positive control), 

 

134 

six days after infiltration. Empty A. tumefaciens was used as negative control. Red and blue 

dashed circles represent infiltrations/co-infiltration for IPI-O1 and IPI-04, respectively. 

 

ii. 

RB-SP/RB-transgenic potato lines 

To further confirm the previous results, RB-SP and RB transgenic potato lines were generated to 

test whether or not these genes interact with the IPI-O effectors in vivo. Several transformation 

events were recovered for each RB gene construct (Table 4.2). All transformation events 

revealed the integration of the NPTII gene based upon PCR analysis and agarose gel detection, 

with a transformation efficiency of 26.6% for RB-SP and 15.5% for RB (Table 4.2). Nine and 

seven transformation events for RB-SP and RB transgenic lines respectively, were discarded due 

to phenotypic abnormalities. 

 

Table 4.2. Transformation efficiency for RB-transgenic lines. 

Construct 

Num. 

Explants 

N. Trans. 
Events** 

RB-SP* 
RB 

110 
97 

26 
17 

Transformation 

Efficiency 
 (%)*** 

Selected 
transgenic 
plants  

26.6 
15.5 

8 
10 

*Synthetic RB gene with a CC domain from S. pinnatisectum and NLR domain core from S. bulbocastanum. 
** Number of transformation events. 
*** Calculated as the percentage of transformation events with NPTII integration. 
 

RT-PCR was conducted to confirm the expression of RB-SP and RB in selected transgenic lines. 

One transgenic line expressing RB-SP was detected (PNT-22, Figure 4.2A) but no expression 

was observed in the remaining lines despite having RB-SP or RB integrated (Figure 4.2A). An 

agroinfiltration assessment was conducted to confirm whether HR is elicited after challenging 

with IPI-O1 and IPI-O4 on PNT-22. HR was triggered using IPI-O1 as expected (Figure 4.2B). 

However, a weak HR was observed when IPI-O4 was infiltrated. 

 

135 

 

Figure 4.2. Agroinfiltration of IPI-O effectors and RB-SP/RB insertion and expression in DRH-

S5 transgenic potato lines. (a). RB-SP detection and expression within selected transgenic lines. 

NPTII and SUS3 were used to confirm RB-SP integration and as an internal control for RT-PCR, 

respectively. NC: Negative control. WT: Wild type. (b) A DRH-S5 transgenic line (PNT-22) 

carrying the RB-SP synthetic gene revealed HR after challenging with both IPI-O effectors. A 

weak HR-type of response is observed on negative control caused by mechanical damage. 

 

136 

iii. 

Coiled-Coil homodimerization disruption mediated by IPI-O4 

A Y2H assay was conducted to validate the CC dimerization suppression when IPI-O4 is 

present. A weak to null self-association was observed between CC domains from blb and pnt 

respectively (Figure 4.3). These results were consistent across each technical replicate and only 

the positive control presented a strong interaction in both DDO/X/A and QDO/X/A dropout 

media. Due to this lack of interaction, we were unable to determine the role IPI-O4 plays in CC 

dimerization. 

Figure 4.3. Coiled-Coil (CC) domain dimerization assay using yeast-two-hybrid. Representative 

CC-domain self -associations are shown for either S. bulbocastanum (blb) and S. pinnatisectum 

(pnt) CC domains. Three independent yeast transformations presented weak to null growth when 

 

compared with a positive control. 

 
V. Discussion 

Plant disease resistance relies on the ability to recognize pathogen-associated signatures through 

specialized plant receptors. Plant breeding programs have leveraged this detection system to 

introduce multi-pathogen resistance into new cultivars, significantly reducing crop production 

losses caused by plant diseases (Gupta et al., 2015). However, much of the research conducted to 

 

137 

date has been restricted to investigations of host-pathogen interactions in vitro, limiting the 

recognition of additional factors contributing to disease resistance responses (Gupta et al., 2015). 

In this study, in vitro and in vivo approaches were used to elucidate the interaction of potato CC 

domains from a synthetic RB gene and the late blight IPI-O effectors. 

The impact of IPI-O1 and IPI-O4 effectors over the RB-mediated resistance in vivo was 

accessed using transient and stable RB expression. RB and IPI-O1 co-infiltration in N. 

benthamiana leaves elicited HR (Figure 4.1), supporting in vitro and in vivo observations 

reported by Liu and Halterman (2009), Halterman et al. (2010), and Chen et al. (2012). 

However, HR was observed when IPI-O1 and IPI-O4 effectors were infiltrated in a transgenic 

line expressing the synthetic RB-SP gene (Figure 4.2). Although some technical replicates did 

not elicit a strong HR when IPI-O4 was infiltrated, these results suggest that additional 

mechanisms may contribute to HR when RB-SP is present in potato. Moreover, these findings, 

also indicate that IPI-O1 may not be the sole factor for triggering HR. Further analysis should be 

conducted using more transgenic lines expressing RB-SP to confirm these results. 

Interestingly, all transformation events subjected to PCR amplified an RB-SP or RB band. 

However, most of them did not express these genes as revealed by RT-PCR (Figure 4.2A). 

Previous studies have shown a similar finding, associated mainly to transgene silencing 

mechanisms in plants subjected to transgenesis (Fagard and Vaucheret, 2000; Finnegan and 

McElroy, 1994). In particular, transgene co-suppression, correlated with a high gene copy 

number, but also DNA methylation, have been reported as the principal causes of transcriptional 

repression (Fagard and Vaucheret, 2000; Finnegan and McElroy, 1994). Thus, RB 

expression/translation may be suppressed by internal silencing cell machinery in DRH-S5. 

 

138 

Chen et al. (2012) reported no physical interactions between the CC domains from either blb 

or pnt, and the IPI-O4 effector using Y2H. In the present study, no HR was observed when RB-

SP and RB were independently co-infiltrated with IPI-O4, confirming that IPI-O4 does not 

interact with the CC domain in N. benthamiana. Interestingly, unlike most RB-SP and RB 

transgenic lines, the PNT-22 line expressed RB-SP and triggered an HR after challenging with 

both IPI-O effectors (Figure 4.2B). These results suggest that other proteins in addition to RB-SP 

may be involved in HR in potato. Although IPI-O1 and IPI-O4 do not exhibit physical 

interactions to the CC domain from pnt in vitro (Chen et al., 2012), it is possible, that undetected 

specific posttranscriptional and posttranslational changes, such as protein folding, must be 

required to interact with RB-SP in vivo. However, with nearly null RB and RB-SP expression in 

transgenic lines, caution must be applied, and further analysis should be conducted to elucidate 

these interactions in vivo. For instance, a detached leaf bioassay using the US-23 P. infestans 

strain could be conducted to confirm these results. 

These findings raise intriguing questions regarding the nature and extent of the RB-mediated 

resistance through the recognition of IPI-O1 and consequent CC dimerization. The present study 

could not validate the interaction with CC domains from either blb or pnt in vitro. Despite 

presenting weak yeast growth on DDO/X/A and QDO/X/A dropout media (comparable with 

negative control), these results are not conclusive. 

Further research should be undertaken to investigate the RB-mediated resistance suppression 

by the IPI-O4 effector from late blight in transgenic potato lines. Although in vitro and in vivo 

observations agree with previous studies using transient expression in N. benthamiana, only one 

line carrying a synthetic RB gene was able to trigger an HR upon IPI-O effectors infiltration. A 

previously undetected interaction was also observed (RB-SP/IPI-O4) suggesting that additional 

 

139 

proteins or recognition mechanism may be involved in an HR in transgenic potato. Moreover, we 

were unable to confirm CC-self associations and therefore further analysis should be conducted 

to confirm these findings. 

 

 

 

140 

LITERATURE CITED 

 

141 

 
 
 
Ahn, Y. K., and Park, T.-H. (2013). Resistance to common scab developed by somatic hybrids 

between Solanum brevidens and Solanum tuberosum. Acta Agriculturae Scandinavica, 
Section B - Soil & Plant Science 63:595–603. 

 
Chen, Y., and Halterman, D. A. (2017). Phytophthora infestans effectors IPI-O1 and IPI-O4 each 

contribute to pathogen virulence. Phytopathology 107:600–606. 

 
Chen, Y., Liu, Z., and Halterman, D. a. (2012). Molecular determinants of resistance activation 

and suppression by Phytophthora infestans effector IPI-O. PLoS Pathogens 8:1–10. 

 
Colton, L. M., Groza, H. I., Wielgus, S. M., and Jìang, J. (2006). Marker-assisted selection for 
the broad-spectrum potato late blight resistance conferred by gene RB derived from a wild 
potato species. Crop Science 46: 589–594. 

 
Devaux, A., Kromann, P., and Ortiz, O. (2014). Potatoes for sustainable global food security. 

Potato Research 57: 185–199. 

 
Fagard, M., and Vaucheret, H. (2000). (Trans)gene silencing in plants: How many mechanisms? 

Annual Review of Plant Physiology and Plant Molecular Biology 51: 167–194. 

 
FAOSTAT (2019). FAOSTAT database collections. Food and Agriculture Organization of the 

United Nations. Available at: http://www.fao.org/faostat/ [Accessed January 27, 2019]. 

 
Finnegan, J., and McElroy, D. (1994). Transgene inactivation: plants fight back!. 

Bio/Technology 12: 883–888. 

 
Friedman, M. (1997). Chemistry, biochemistry, and dietary role of potato polyphenols: A 

Review. Journal of Agricultural and Food Chemistry 45: 1523–1540. 

 
Gupta, R., Lee, S. E., Agrawal, G. K., Rakwal, R., Park, S., Wang, Y., et al. (2015). 

Understanding the plant-pathogen interactions in the context of proteomics-generated 
apoplastic proteins inventory. Frontiers in plant science 6, 352. 

 
Halterman, D. A., Chen, Y., Sopee, J., Berduo-Sandoval, J., and Sánchez-Pérez, A. (2010). 

Competition between Phytophthora infestans effectors leads to increased aggressiveness on 
plants containing broad-spectrum late blight resistance. PloS One 5(5):e10536. 

 
Halterman, D. a, Kramer, L. C., Wielgus, S., and Jiang, J. (2008). Performance of transgenic 

potato containing the late blight resistance gene RB. Plant Disease 92: 339–343. 

LITERATURE CITED 

 

 

142 

Haverkort, A. J., Struik, P. C., Visser, R. G. F., and Jacobsen, E. (2009). Applied biotechnology 
to bombat bate blight in potato aused by Phytophthora Infestans. Potato Research 52: 249–
264. 

 
Koncz, C., Martini, N., Szabados, L., Hrouda, M., Bachmair, A., and Schell, J. (1994). 

“Specialized vectors for gene tagging and expression studies,” in Plant Molecular Biology 
Manual, eds. S. B. Gelvin and R. A. Schilperoort (Dordrecht: Springer Netherlands), 53–74. 

 
Kuhl, J. C., Zarka, K., Coombs, J., Kirk, W. W., and Douches, D. S. (2007). Late blight 

resistance of RB transgenic potato lines. Journal of the American Society for Horticultural 
Science 132: 783–789. 

 
Leesutthiphonchai, W., Vu, A. L., Ah-Fong, A. M. V, and Judelson, H. S. (2018). How does 
Phytophthora infestans evade control efforts?. Modern insight into the late blight disease. 
Phytopathology 108: 916–924. 

 
Li, W., Zarka, K. A., Douches, D. S., Coombs, J. J., Pett, W. L., and Grafius, E. J. (1999). 

Coexpression of potato PVYo coat protein and cryV-Bt genes in potato. J. Amer. Soc. Hort. 
Sci. 124: 218–223. 

 
Liu, Z., and Halterman, D. (2009). Analysis of proteins differentially accumulated during potato 
late blight resistance mediated by the RB resistance gene. Physiological and Molecular Plant 
Pathology 74: 151–160. 

 
Lozano, R., Ponce, O., Ramirez, M., Mostajo, N., and Orjeda, G. (2012). Genome-Wide 

Identification and Mapping of NBS-Encoding Resistance Genes in Solanum tuberosum 
Group Phureja. PloS One 7(4):e34775. 

 
Ma, L., Lukasik, E., Gawehns, F., and Takken, F. L. W. (2012). “The use of agroinfiltration for 

transient expression of plant resistance and fungal effector proteins in Nicotiana 
benthamiana leaves,” in Plant Fungal Pathogens: Methods and Protocols, eds. M. D. Bolton 
and B. P. H. J. Thomma (Totowa, NJ: Humana Press), 61–74. 

 
Murashige, T., and Skoog, F. (1962). A revised medium for rapid growth and bio assays with 

tobacco tissue cultures. Physiol. Plant 15: 473–497. 

 
Nowicki, M., Foolad, M. R., Nowakowska, M., and Kozik, E. U. (2011). Potato and tomato late 
blight caused by Phytophthora infestans: An overview of pathology and resistance breeding. 
Plant Disease 96: 4–17. 

 
Pomerantz, A., Cohen, Y., Shufan, E., Ben-Naim, Y., Mordechai, S., Salman, A., et al. (2014). 

Characterization of Phytophthora infestans resistance to mefenoxam using FTIR 
spectroscopy. Journal of Photochemistry and Photobiology B: Biology 141: 308–314. 

 
Raymundo, R., Asseng, S., Robertson, R., Petsakos, A., Hoogenboom, G., Quiroz, R., et al. 
(2017). Climate change impact on global potato production. Eur. J. Agron. 100: 87-98. 

 

143 

 
Rodewald, J., and Trognitz, B. (2013). Solanum resistance genes against Phytophthora infestans 

and their corresponding avirulence genes. Molecular Plant Pathology 14: 740–757. 

 
Smart, C. D., and Fry, W. E. (2001). Invasions by the late blight pathogen: Renewed sex and 

enhanced fitness. Biological Invasions 3: 235–243. 

 
Song, J., Bradeen, J. M., Naess, S. K., Raasch, J. a, Wielgus, S. M., Haberlach, G. T., et al. 

(2003). Gene RB cloned from Solanum bulbocastanum confers broad spectrum resistance to 
potato late blight. Proceedings of the National Academy of Sciences of the United States of 
America 100: 9128–9133. 

 
Sparks, A. H., Forbes, G. A., Hijmans, R. J., and Garrett, K. A. (2014). Climate change may 
have limited effect on global risk of potato late blight. Global Change Biology 20: 3621–
3631. 

 
The Potato Genome Sequencing Consortium (2011). Genome sequence and analysis of the tuber 

crop potato. Nature 475: 189–195. 

 
Vinson, J. a., Demkosky, C. a., Navarre, D. a., and Smyda, M. a. (2012). High-antioxidant 

potatoes: Acute in vivo antioxidant source and hypotensive agent in humans after 
supplementation to hypertensive subjects. Journal of Agricultural and Food Chemistry 60: 
6749–6754. 

 
Xing, S., Wallmeroth, N., Berendzen, K. W., and Grefen, C. (2016). Techniques for the analysis 

of protein-protein interactions in vivo. Plant physiology 171: 727–758. 

 
 

 

 

144 

Chapter 5  

GENERAL CONCLUSIONS 

I. 

Improving pathogen-resistance in potato using genomic selection 

The genomic selection study described here uses whole genome-regressions to access the genetic 

architecture and prediction accuracy for late blight and common scab resistance in a breeding 

population from the Michigan State University (MSU) potato breeding program. In this study, 

high-quality Single Nucleotide Polymorphisms (SNPs) and field data collected during seven and 

nine years for late blight and common scab were used. These data reveal a moderately high 

genomic heritability value for late blight (0.46±0.04) and common scab (0.45±0.017). 

The extent of genotype-by-year interaction was high for late blight and low for common 

scab, reflecting high disease pressures in the field caused by changes of pathogen aggressiveness 

and late blight strain prevalence across years. In particular, for late blight, we revealed that 

multiple SNPs contribute to late blight resistance on previously reported resistance hotspot 

regions. These results uncover the disease resistance introgression history of our germplasm 

collection coming from species such as S. demissum and S. berthaultii (Massa et al. 2015; 

Manrique-Carpintero, personal communication). Unlike late blight, a new locus located in a 

WRKY transcription factor, with a sizable contribution to inter-individual differences, was 

detected for common scab. This SNP along with those with larger-SNP variance detected in this 

study can be used to support marker-assisted selection programs using genotyping technologies 

such as KASP, representing the first attempt to date for early selection of common scab 

resistance lines. 

Prediction accuracy assessments demonstrated the applicability of genomic prediction for 

tetraploid potato breeding. For both traits, more than 90% of the genetic variance could be 

 

145 

captured with an additive model. For late blight, small but statistically significant gains in 

prediction accuracy were obtained using a model that accounted for both additive and dominance 

effects. For common scab, the highest prediction accuracy was achieved using an additive model. 

Further efforts to increase prediction accuracies for late blight and common scab resistance 

should consider expanding sample size and marker density and also integrate this breeding 

strategy into the MSU potato breeding scheme. In particular, the development of a 35K SNP 

array will not only allow increased marker density but also reduce the ascertainment bias by 

including new and rare genotypic variants. Additionally, the models evaluated in this work can 

also be extended and integrated into current diploid breeding strategies. 

 

II. 

Contributing to diploid potato breeding using targeted mutagenesis 

The targeted mutagenesis study described here uses the CRISPR/Cas9 system to overcome self-

incompatibility (SI) in diploid potato. We first computationally identified three new S-RNase 

alleles, with flower-specific expression, in two diploid SI potato lines (DRH-195 and DRH-310). 

These alleles presented high inter-restricted amino-acid similarities to reported Solanaceae S-

RNase sequences, supporting the theory that S-RNase divergence precedes speciation (Ioerger et 

al., 1990). 

An RH-specific S-RNase allele was mapped to chromosome I within the peri-centromeric 

region, consistent with the localization of the S-locus in a low recombination region, favoring 

outcrossing. S-RNase knock-out (KO) lines were obtained using a dual single-guide RNA 

(sgRNA) strategy, generating premature stop codons by targeting conserved regions on each S-

RNase exon. After self-pollination, fruits were set in selected KO lines in T0 and T1, 

demonstrating a stable S-RNase KO transmission. Moreover, Cas9-free KO lines were also 

 

146 

detected in T1 lines, representing a significant advance towards the generation of transgene-free 

self-compatible (SC) potato cultivars that may not be regulated by the United States Department 

of Agriculture, Animal and Plant Health Inspection Service (USDA APHIS). 

Plasticity in SI was also observed in wild-type (WT) plants after self-pollination. These 

results demonstrated that besides the S-RNase-based SI system, other modifier loci can modulate 

pollen rejection in gametophytic SI, as reported in several Solanum species (Goldraij et al., 2006; 

McClure et al., 1999; O’Brien et al., 2002). In particular, proteins such as eEF1A or 120K, can 

directly interact with S-RNase, contributing to the strength of the SI response (Goldraij et al., 

2006; Soulard et al., 2014). For instance, the suppression of pollen tube inhibition has been 

achieved targeting a gene encoding the 120K protein via RNA interference (Liu et al., 2009). 

Therefore, knocking-out these modifier loci along with S-RNase may introduce a consistent and 

stable self-compatibility in diploid potato. 

Environmental factors also play an essential role in SI plasticity. For instance, temperature 

fluctuations, photoperiod, glucose starvation, and humidity significantly reduced SI in species 

such as S. peruvianum after self-pollination (Webb and Williams, 1988). To further explore 

whether conditions such as temperature contribute to the strength of the SI response in WT and 

S-RNAse KO lines, in vivo and in vitro analysis could be performed. As reported in citrus 

(Distefano et al., 2012), a species under gametophytic SI, self-pollinated plants and excised 

flowers can be placed in a growth chamber under different temperature conditions (i.e., 20, 25 

and 30 ˚C). The effect of temperature on pollen germination and pollen tube growth can be 

observed under microscope after style staining. This approach could provide insight on the 

impact of specific environmental conditions over SI plasticity and may be considered as an 

alternative strategy for the generation of SC diploid potato. Moreover, heritability estimates can 

 

147 

be accessed to investigate the genetic and environmental component explaining self-

compatibility in WT and KO lines. 

Overall, additional work must be conducted to increase the strength and stability of the SC 

phenotype. Targeting primary fertility-related genes including S-RNase and non-stylar factors 

will contribute to incorporate self-compatibility into diploid potato inbred lines. 

 

III. 

Understanding late blight resistance suppression 

The validation of the RB-mediated resistance suppression in potato assessment described here 

uses in vivo and in vitro approaches to confirm this suppression. Using a heterologous system, 

we confirmed the interaction between the Coiled-Coil (CC) domain from the resistance to the 

late blight (RB) protein and the late blight IPI-O1 effector leading to a hypersensitive responses 

(HR). 

Diploid transgenic potato lines were generated using a synthetic RB gene carrying a CC 

domain from S. pinnatisectum and an NLR domain core from S. bulbocastanum (RB-SP), using 

an RB gene as a control. The PNT-22 transgenic line, expressing the RB-SP gene was selected 

for agroinfiltration assays using the IPI-O1 and IPI-O4 effectors. The HR was elicited for both 

late blight effectors, suggesting that additional mechanisms may be involved in the resistance 

response to late blight in transgenic diploid potato. Further analyses should be conducted to 

explore whether proteins other than RB-SP or unreported interactions between the IPI-O 

effectors and specific domains from RB-SP, contribute to HR. Moreover, additional transgenic 

lines expressing RB-SP should be generated to confirm the observed phenotype and determine a 

possible relationship between HR strength and RB-SP copy number. Likewise, a detached-leaf 

 

148 

bioassay using the US-23 late blight genotype will also provide a broader comprehension of the 

RB-SP-mediated resistance in diploid potato. 

This study was unable to demonstrate CC-self association using a yeast-two-hybrid system 

(Y2H), and therefore we could not validate the IPI-O4 suppression role over the RB-mediated 

resistance. Despite observing a weak yeast growth in selection media for using CC domains for 

both RB genes, the results were not conclusive. To confirm whether these results were a product 

of bait toxicity, further analysis should be conducted including yeast co-transformations using 

empty bait plasmids along with CC. Similarly, to determine whether the lack of self-association 

was due to a lack of CC expression, western blot assays should be conducted to confirm this 

hypothesis. Similarly, additional work should be conducted to test if the lack of CC-interaction 

was a result of a loss of yeast glycosylation signatures related to posttranscriptional modification 

as suggested by Xing et al. (2016) for Y2H systems. 

 

IV. 

Final considerations 

Advances in crop-based breeding have enabled the identification of thousands of genomic 

variants that can be harnessed for selection of superior genotypes without multiple phenotypic 

cycles through genomic selection (GS) (Hu et al., 2018). However, multi-trait GS is restricted to 

highly correlated traits (Jia and Jannink, 2012), which underscores the need for integrating 

genomic technologies towards the improvement of multiple related or unrelated traits to respond 

rapidly to increasing food demand. In particular, genome editing technologies could contribute to 

the generation of superior breeding materials by targeting genomic regions associated with 

deleterious genes or enhancing the expression of important agronomic-related traits. 

 

149 

Recently, a targeted recombination strategy, combining GS and genome editing tools, was 

proposed to increase genetic gains for several crops. This approach involves the use of 

homologous recombination or crossing-over induction on genomic regions where marker effects 

have been estimated (Bernardo, 2017). An increase of genetic gains could be achieved for those 

lines in which chromosome segments with desirable allele combinations are present (Bernardo, 

2017). Despite the inability to target all regions with high estimated markers effects, simulation 

analysis on major crops such as maize, soybean, wheat, barley, and pea (Bernardo, 2017; Ru and 

Bernardo, 2018), have demonstrated the feasibility of targeted recombination, directed to reduce 

the breeding cycle in crops like potato. 

A constant improvement in potato genotyping platforms, including sequencing initiatives 

such as the potato pan-genome, will enable the integration of genomics tools for potato 

improvement. The recognition of new causative genetic variants for agronomic-related traits 

using genome wide-association and GS analysis will become a primary input for targeted 

recombination in potato. This new breeding method may allow the generation of novel allele 

combinations in potato breeding, translating genomic and data modeling into genetic gains in the 

field. 

 

 

 

 

 

 

 

 

150 

LITERATURE CITED 

 

151 

 
Bernardo, R. (2017). Prospective targeted recombination and genetic gains for quantitative traits 

in maize. The plant genome 10: 1–9. 

 
Distefano, G., Hedhly, A., Las Casas, G., La Malfa, S., Herrero, M., and Gentile, A. (2012). 
Male–female interaction and temperature variation affect pollen performance in Citrus. 
Scientia Horticulturae 140: 1–7. 

 
Goldraij, A., Kondo, K., Lee, C. B., Hancock, C. N., Sivaguru, M., Vazquez-Santana, S., et al. 

(2006). Compartmentalization of S-RNase and HT-B degradation in self-incompatible 
Nicotiana. Nature 439(7078): 805-10. 

 
Hu, H., Scheben, A., and Edwards, D. (2018). Advances in integrating genomics and 

bioinformatics in the plant breeding pipeline. Agriculture 8, 75. 

 
Ioerger, T. R., Clark, A. G., and Kao, T. H. (1990). Polymorphism at the self-incompatibility 
locus in Solanaceae predates speciation. Proceedings of the National Academy of Sciences 
87: 9732 LP – 9735. 

 
Jansky, S. H., Charkowski, A. O., Douches, D. S., Gusmini, G., Richael, C., Bethke, P. C., et al. 

(2016). Reinventing potato as a diploid inbred line–based crop. Crop Science 56: 1412–1422.  

Jia, Y., and Jannink, J.-L. (2012). Multiple-trait genomic selection methods increase genetic 

value prediction accuracy. Genetics 192: 1513–1522. 

 
Liu, B., Morse, D., and Cappadocia, M. (2009). Compatible pollinations in Solanum chacoense 

decrease both S-RNase and S-RNase mRNA. PloS One 3;4(6):e5774. 

 
Massa, A. N., Manrique-Carpintero, N. C., Coombs, J. J., Zarka, D. G., Boone, A. E., Kirk, W. 

W., et al. (2015). Genetic linkage mapping of economically important traits in cultivated 
tetraploid potato (Solanum tuberosum L.). G3: Genes, Genomes, Genetics 5: 2357–2364. 

 
McClure, B., Mou, B., Canevascini, S., and Bernatzky, R. (1999). A small asparagine-rich 
protein required for S-allele-specific pollen rejection in Nicotiana. Proceedings of the 
National Academy of Sciences. 96: 13548–13553. 

 
O’Brien, M., Kapfer, C., Major, G., Laurin, M., Bertrand, C., Kondo, K., et al. (2002). Molecular 

analysis of the stylar-expressed Solanum chacoense small asparagine-rich protein family 
related to the HT modifier of gametophytic self-incompatibility in Nicotiana. Plant J 32: 
985–996. 

 
Ru, S., and Bernardo, R. (2018). Targeted recombination to increase genetic gain in self-

pollinated species. Theoretical and Applied Genetics, 132(2):289-300. 

LITERATURE CITED 

 
 

 

 

152 

Soulard, J., Boivin, N., Morse, D., and Cappadocia, M. (2014). eEF1A is an S-RNase binding 

factor in self-incompatible Solanum chacoense. PLOS ONE 9(2): e90206. 

 
Webb, M. C., and Williams, E. G. (1988). Effects of temperature, light, nutrients and carbon 

dioxide on the strength of the self-incompatibility response in detached flowers of 
Lycopersicon peruvianum. Ann. Bot. 61: 395–404. 

 
Xing, S., Wallmeroth, N., Berendzen, K. W., and Grefen, C. (2016). Techniques for the analysis 

of protein-protein interactions in vivo. Plant physiology 171: 727–758. 

 
 

 

153