Gladiolus Breeding for Rapid Generation Cycling for Potted Plant

Home , Crocoideae, Crocosmia, Crocosmia aurea, Freesia alba, Freesia laxa, Gladiolus

Gladiolus breeding for rapid generation cycling for potted plant production and the discovery of gladiolus genes, UPSTREAM OF FLOWERING LOCUS C (UFC) and FLOWERING LOCUS C

EXPRESSOR (FLX)

A DISSERTATION SUBMITTED TO THE FACULTY OF THE UNIVERSITY OF MINNESOTA BY

Jaser A S A Aljaser

IN PARTIAL FULFILMENT OF THE REQUIRMENTS FOR THE DEGREE DOCTOR OF PHILOSOPHY

Neil O. Anderson, Advisor

July 2020

Acknowledgments

I would like to give many thanks and acknowledgements for the following people who helped me in the research project and guided me through experiments:

• Dr. Neil Anderson (Primary Investigator) for allowing me to work in this project, for his

overall guidance during the whole research, providing materials, equipment, comments

on the dissertation and encouragement in graduate school.

• Dr. Alan Smith, Dr. Neil Olszewski and Dr. Changbin Chen (Committee) for their

guidance in experiments, advise and commenting on the dissertation.

• Dr. Andrzej Noyszewski for his help in the lab experiments procedures and guidance in

the lab analyzing the DNA results.

• Dr. Neil Anderson’s lab members for their help in providing greenhouse suitable

environment for gladiolus cultivation.

Dedication

• Kuwait University for offering me the scholarship and taking care of the financial support

for my graduate studies in horticultural at the University of Minnesota.

• My parents and siblings for their love, encouragement and support to help me overcome

the challenges.

Jaser Aljaser

Abstract

Gladiolus (Gladiolus ×hybridus Rodigas) is a geophytic floriculture crop, cultivated for cut flower and garden ornamental uses. It is a perennial crop and is characterized to have a long juvenile period of up to 5 years, with tall plant height which is undesirable for potted production and poorly understood in flowering pathway. In this research, the constant selection in gladiolus breeding at the University of Minnesota for rapid generation cycling (RGC) resulted in reducing the juvenility time from 3 – 5 years into a range of 3 – 11 months from germinating seedlings, recurrent selection breeding method was significantly early in flowering time (21.63 weeks).

Also, our gladiolus genotypes have reduced dormancy, are able to re-sprout after senescence, which is a new trait in gladiolus. These traits in RGC genotypes were tested further with the application of ancymidol, a gibberellin inhibitor to control height in potted plants. The high concentrations of ancymidol can inhibit flowering for commercial cultivars. However, the RGC lines had significant reductions in plant height to a more acceptable plant size within the aesthetic ratio and the ability to flower while commercial cultivar comparisons failed to flower at higher concentrations of ancymidol. All this knowledge drove us to test for the existence of flowering genes in gladiolus. Herein we were able to discover the existence of UPSTREAM OF

FLOWERING LOCUS C (UFC) and FLOWERING LOCUS C EXPRESSOR (FLX) in gladiolus.

UFC is an adjacent gene to FLOWERING LOCUS C (FLC), the floral repressor in Arabidopsis thaliana, while FLX gene upregulates FRIGIDA (FRI) which upregulates FLC expression. The amino acid sequence of UFC has up to 57% identity to Musa acuminata. The discovered FLX gene in gladiolus is partially completed which is missing 2 exons and shows up ~65% of identity of FLX to Ananas comosus. These two, newly discovered genes in gladiolus, can provide better

iii

understanding of the flowering and vernalization response in ornamental geophytes and begin clarification of the effects of breeding for RGC on flowering genes / alleles as well as the application of plant growth retardants for production of dwarf gladiolus for potted plant production

Table of contents

Acknowledgements ………………………………………………………………………………. i

Dedication ………………………………………………………………………………………... ii

Abstract …………………………………………………………………………………………. iii

Table of contents …………………………………………………………………………………. v

List of tables ……………………………………………………………………………………. vii

List of figures ……………………………………………………………………………………. ix

List of Appendices …………………………………………………………………………….... xii

1.0 Literature review ………………………………………………………………………… 1

1.1 Gladiolus, Taxonomy and distribution ……………………………………………… 1

1.2 Botanical description ………………………………….…………….……….……… 2

1.3 Crop Production and sales …………………………………………………….….…. 3

1.4 Pests and diseases …………………………………………………………………… 4

1.5 Gladiolus breeding ………………………………………………………………….. 5

1.6 Gladiolus genetics …………………………………………………………………... 7

1.7 Gladiolus and the use of plant growth regulators …………………………………… 8

1.8 Flowering models and genes ………………………………………………………. 10

1.9 Literature cited ……………………………………………………………….….… 14

2.0 Gladiolus breeding for rapid generation cycling and reduced dormancy ………….…... 31

2.1 Preface ………………………………………………………………………….… 32

2.2 Introduction ……………………………………………………………….….….… 33

2.3 Material and Method ………………………………………………………….…… 37

2.4 Results ………………………………………………………………….….….…… 39 v

2.5 Discussion ……………………………………………………………………….… 41

2.6 Literature cited ……………………………………………………………….….… 46

3.0 Effects of a gibberellin inhibitor on flowering, vegetative propagation and production of rapid generation cycling gladiolus for potted plant production ………………………... 69

3.1 Preface ……………………………………………………………………….….… 70

3.2 Introduction ……………………………………………………………….….….… 71

3.3 Material and Methods ……………………………………………………………… 74

3.4 Results …………………………………………………………………….….….… 77

3.5 Discussion ……………………………………………………………………….… 78

3.6 Literature cited ……………………………………………………………….….… 82

4.0 Discovery of UPSTREAM OF FLOWERING LOCUS C (UFC) and FLOWERING LOCUS C EXPRESSOR (FLX) Genes in Gladiolus ×hybridus and G. dalenii ………... 96

4.1 Preface ………………………………………………………………….………… 97

4.2 Introduction ……………………………………………………………….….….… 98

4.3 Material and Method ……………………………………………………………... 103

4.4 Results …………………………………………………………………….……… 107

4.5 Discussion ………………………………………………………………………... 109

4.6 Literature cited …………………………………………………………………… 114

5.0 General conclusion …………………………………………………………………… 138

5.1 Literature cited …………………………………………………………………… 141

Appendices …………………………………………………………………………….……… 143

List of tables

Chapter 2.0 Gladiolus breeding for rapid generation cycling and reduced dormancy

Table 1. Mean number of weeks of gladiolus seed germination, number of weeks of flowering from seed germination and percentage of flowering in every generation ………………………. 50

Table 2. Number of gladiolus seedling, number of re-sprouted seedling, dormant and percentage of re-sprouted gladiolus plants in every generation and 1:1 Chi-square (χ2) test ratios of the tested

Gladiolus generations …………………………………………………………………………... 51

Chapter 3.0 Effects of a gibberellin inhibitor on flowering, vegetative propagation and production of rapid generation cycling gladiolus for potted plant production

Table 1. Number of replicates / treatment (0, 100, 400 ppm A-Rest®) and commercial or breeding source of the tested Gladiolus genotypes. …………………………………………….. 87

Table 2. Influence of A-Rest® concentrations (0, 100, and 400 ppm) on the number of flowering plants (frequency of flowering) and 1:1:1 Chi-square (χ2) test ratios of the tested Gladiolus genotypes .………………………………………………………………………………………. 88

Table 3. Influence of A-Rest® concentrations (0, 100, and 400 ppm) on the number of flowering plants (frequency of flowering) and mean number of weeks reached to flowering in each genotypes of the tested Gladiolus genotypes …………………………………………………... 89

Table 4. Mean plant height (cm), flower stalk height (cm), number of stalks, leaf width (cm), no. of corms, fresh weight (FW) of corms (g), number of cormels, and fresh weight (FW) of cormels

(g) for seven gladiolus genotypes corms treated with different concentrations of A-Rest® (0, 100, and 400 ppm) …………………………………………………………………………………… 91

vii

Chapter 4.0 Discovery of UPSTREAM OF FLOWERING LOCUS C (UFC) and FLOWERING

LOCUS C EXPRESSOR (FLX) Genes in Gladiolus ×hybridus and G. dalenii

Table 1. The Gladiolus genotypes used for the study …………………………………………. 125

Table 2. The identity of amino acid sequences and number (%) of two UFC proteins (GhUFC-A,

GhUFC-B) in two Gladiolus (genotypes 1 and 15) in relation to other species (Gene locus/ID) through pair alignment ………………………………………………………………………… 126

Table 3. Identity of amino acid sequences, number (%) of UFC proteins in the conserved domain

DUF966 in Gladiolus genotypes in relation to other species through pair alignment ………… 127

Table 4. Number (%) of amino acid sequences of GhFLX protein in Gladiolus genotypes 3 and 6

(genotype 16 is identical to genotype 6) in relation to the other species through pair alignment

………………………………………………………………………………………………….. 128

viii

List of figures

Chapter 2.0 Gladiolus breeding for rapid generation cycling and reduced dormancy

Figure. 1 A. University of Minnesota gladiolus breeding lines with unique flower colors and patterning ……………………………………………………………………………………….. 52

Figure 2. The breeding scheme for rapid generation cycling (RGC) gladiolus ………………… 53

Figure 3. The University of Minnesota gladiolus breeding lines have large capsule (fruit) as evidence of fertility ……………………………………………………………………………... 54

Figure 4. Representation of number of flowering plants in cycles (years) in respect of parent generation (Green line) and RGC-1 population (Blue line) …………………………………….. 55

Figure 5. Phenotypes of the rapid generation cycling (RGC-1) gladioli ……………………….. 56

Figure 6. Frequency of the number of flowering plants in the RGC-2 recurrent selection generation ……………………………………………………………………………………….. 57

Figure 7. Phenotypes of the RGC-2 recurrent selection gladioli ……………………………….. 58

Figure 8. The number of gladiolus plants in rapid generation cycling (RGC-3) inbreds flowering in cycles 1 and 2 ………………………………………………………………………………… 59

Figure 9. Phenotypes of rapid generation cycling (RGC-3) ……………………………………. 60

Figure 10. A. Phenotypes of the fastest flowering gladiolus ever reported (seed to flower in 12 weeks) in Year 1 and replanted in following season (Year 2) ………………………………….. 61

Figure 11. Phenotypes of a rapid generation cycling (RGC-1) plant in its cycle 1 (Year 1) and the same genotype planted in second year (Year 2) ………………………………………………... 62

Figure 12. Phenotype of a gladiolus seedling which flowered from a small cormel (2 cm in circumference) without a cooling treatment ……………………………………………………. 63

Figure 13. Wild type gladiolus having 3 anthers and 3 stigmas while some Rapid generation cycling (RGC) have different number of anthers and stigmas ………………………………….. 64

Figure 14. A. Cultivated gladiolus ‘Glamini’® florets are horizontal in position while rare feature in rapid generation cycling (RGC) gladiolus with an upward floret position …………………... 65

Figure 15. Phenotype of a re-sprouting rapid generation cycling (RGC) gladiolus without a cooling treatment ……………………………………………………………………………….. 66

Figure 16. The morphological differences between seed propagated rapid generation cycling

(RGC) cycle 1gladiolus and a vegetatively propagated cormel ………………………………… 67

Figure 17. Phenotype of a desired ideotype gladiolus for potted production …………………... 68

Chapter 3.0 Effects of a gibberellin inhibitor on flowering, vegetative propagation and production of rapid generation cycling gladiolus for potted plant production

Figure 1. ‘Bananarama’ gladiolus treated with A-Rest® ………………………………………. 93

Figure 2. Rapid Generation Cycling (RGC) RGC-3 gladiolus treated with A-Rest® ………….. 94

Figure 3. Gladiolus treated with A-Rest® concentration of 400 ppm. RGC-2 reached to flowering, while ‘Bananarama’ remained in vegetative state …………………………………... 95

Chapter 4.0 Discovery of UPSTREAM OF FLOWERING LOCUS C (UFC) and FLOWERING

LOCUS C EXPRESSOR (FLX) Genes in Gladiolus ×hybridus and G. dalenii

Figure 1. Model represents portion of flowering pathway regarding the role of FLX gene in flowering along with UFC gene in Arabidopsis and temperate dicots, and monocots ………... 129

Figure 2. Multi-alignment of UFC coding sequence in Gladiolus ×hybridus (GhUFC) and

Gladiolus dalenii (GdUFC) …………………………………………………………………… 130

Figure 3. Multi-alignment of UFC amino acid sequence in Gladiolus ×hybridus (GhUFC) and

Gladiolus dalenii (GdUFC), the alignment is for the 17 genotypes …………………………... 131

Figure 4. Intron-exon configuration of the UFC genes in Gladiolus ×hybridus of genotypes 1 and

15 in relation to several species ……………………………………………………………… 132

Figure 5. Alignment of the globular region containing DUF966 domain of UFC proteins from

Gladiolus ×hybridus of genotypes 1 and 15, Ananas comosus, Musa acuminata, Elaeis guineensis

Asparagus officinalis, Arabidopsis thaliana and Glycine max ………………………………... 134

Figure 6. The phylogenetic tree of all UFC genotypes in Gladiolus ………………………….. 135

Figure 7. Multi-alignment of FLX amino acid sequence in Gladiolus ×hybridus (GhFLX) with other species; Arabidopsis thaliana, Ananas comosus, Elaeis guineensis, Musa acuminata and

Glycine max …………………………………………………………………………………… 136

Figure 8. Pair alignment of FLX protein in Ananas comosus and Gladiolus ×hybridus genotype

16 ………………………………………………………………………………………………. 137

List of Appendices

A1. The primer design for VRN2 gene in the tested cereal and gladiolus samples …………… 143

A2. Gel electrophoresis image of PCR for VRN2 gene bands of gladiolus genotypes and cereals

………………………………………………………………………………………………….. 144

A3. Gel electrophoresis image PCR for VRN2 gene band of Triticum monococum, gladiolus and geophytic genotypes …………………………………………………………………………… 145

A4. The desirable height for Rapid Generation Cycling (RGC) genotypes for gladiolus potted production …………………………………………………………………………………… 146

xii

1.0 Literature review

1.1 Gladiolus, taxonomy and distribution

The genus Gladiolus is member of Iris family (Iridaceae), both the genus and common name is the same and the word “gladiolus” is derived from the Latin word gladius which means

“little sword”, in reference to the leaf shape being lanceolate and, thus, sword-shaped (Goldblatt

& Manning 2008). There are 270 species in the genus Gladiolus (Manning et al 2014) with ~263 species found in sub-Saharan Africa and Madagascar, seven species are native to Eurasia

(Goldblatt & Manning 2008), ~170 species are native in the region of southern Africa while 93 species are from tropical Africa (Manning et al 2014). The centers of origin and diversity for gladiolus is South Africa. There are only three species found in both sub-Saharan (Arabian

Peninsula) and Eurasian: Gladiolus dalenii, G. candidus and G. abyssinicus (Goldblatt 1996). The most geographically distributed gladiolus species is G. dalenii which is found in sub-Saharan

Africa and the Arabian Peninsula; it has three subspecies: G. dalenii subsp. dalenii, G. dalenii subsp. andongensis and G. dalenii subsp. welwitschii (Goldblatt 1996). Gladiolus also known as

“corn flag” because it was a weed in a corn field (corn was the general term for cereals in old word definition). Thus, gladiolus used to grow as a weed in wheat and barley fields in the ancient

Greek and Levant region. The name “corn flag” was given specifically to G. italicus (=G. segetum) because it has pink colored florets, although G. italicus is not that abundant in Italy

(Anderton and Park 1989). Gladiolus italicus is a Eurasian species that can be found in

Mediterranean region, except for Egypt (Peri 2015). It is the most abundant gladiolus species in the Mediterranean region (Sheasby 2007) and extends eastward into Iran, and reportedly grows in

Kuwait (Boulos 1988) and Oman in the Arabian Peninsula (Pickering & Patzelt 2008). The species’ corms were roasted and eaten (Lewis, Obermeyer & Barnard 1972). With this vast

distribution of gladiolus, discoveries of new species are still occurring as recently as 2019

(Manning et al 2019).

1.2 Botanical description

Gladiolus is identified by its corm, a geophyte structure which is an underground, modified swollen stem that has storage properties; starch is the carbohydrate stored in gladiolus’ corms (Yasuda & Yokoyama 1955). Corms are mainly found in monocotyledons. However, there are lesser known dicotyledon species with corm structures, such as Liatrus species (De Hertogh and Le Nard 1993) and Stylidium petiolare (Pate & Dixon 1982). Gladioli are commercially propagated vegetatively with corms; the planted ‘mother’ corm produces daughter corms by the end of the current season’s growing cycle. Daughter corms are attached on top of the ‘mother’ corm with cormels (smaller corms) attached to the bottom of the daughter corms, arising from the basal plate (Cohat 1993). Upon sprouting, the corm produces adventitious roots as well specialized contractile roots, which are thick fleshy roots that pull each corm deeper into the ground (Halevy 1986). Sexual reproduction through flowering in gladiolus occurs via self- and cross-pollinations (since most species are self-compatible) to create inbreds and hybrid selections and cultivars, respectively. Gladiolus inflorescences are arranged in a spike arrangement, each floral spike consists of individual florets; the number of florets per spike varies and can reach up to 40 florets in G. oppositiflorus (Beal and Hottes, 1916). The fruit of the gladiolus is a capsule, while seed can either be winged or wingless (Goldblatt 1996).

1.3 Crop production and sales

Gladiolus production is usually for either cut flower (floral design) purposes or for corm production. Harvesting high quality cut-flower stems and subtending leaves reduces cormel production by 30%, due to lowering the photosynthetic area (“source”) and reducing, thereby, photosynthates available to the “sink” in corms and cormels (Rees 1992). Preserving the number of leaves without cutting them increases the size of corm and cormels (Tomiozzo, 2019).

Vegetative propagation through cormels in gladiolus is an ideal method to preserve genotypes and there are old cultivars still sold commercially such as G. ×colvillei ‘Albus’ from 1872 which was a sport of the very first interspecific gladiolus hybrid, G. ×colvillei released in 1823 (Beal and Hottes, 1923).

The techniques to increase the number of corms and cormels relies on providing sufficient macronutrients of 122.kg/ha of N, 36 kg/ha of P2O5, 257 kg/ha of K2O, 150 kg/ha of

CaO, and 34 kg/ha of MgO for field production (Cohat,1993). Harvested corms and cormels undergo curing and cleaning by leaving corms in the sun for two to three days to dry out the corm in order to reduce pathogen growth (Monge, 1981). This is followed by immersing corms in a fungicide for 15 – 30 minutes to eliminate fungi on corms (The International Flower Bulb Centre,

2010). Corm sizes for commercial sale can vary with “jumbo” being more than 5.1 cm in circumference, “large” size is between 5.1-3.8 cm in circumference, “medium” size between 3.8-

2.8 cm in circumference and “small” is less than 2.8 cm in circumference (Singh, 2006). In addition to vegetative propagation using corms and cormels, gladiolus can be also vegetatively propagated through tissue culture (Prasad & Gupta, 2006). Tissue culture can be efficient, alternative method for vegetative propagation when genotypes lack vigor for production of corms and cormels (Prasad & Gupta, 2006).

Commercial gladiolus cut flower production for the cut flower market and use in floral designing depends on factors affecting the inflorescence and corms, as cooling (vernalization) is required for corms to reduce inhibitors (abscisic acid) whereas long day photoperiods are required for flower bud initiation and development (Cohat, 1993). The floral spike is cut and harvested when petal coloration starts in the lowest floret but before it reaches anthesis (Hertogh, 1996).

Gladiolus floral spikes can be stored in dry conditions at 4oC for 2 to 3 weeks (Dole and Wilkins,

2005). An advantage of gladioli flowers is their low sensitivity to ethylene, thus increasing vase life of the cut flowers (Serek, 1994).

Gladiolus have been in the top 10 cut flowers in Dutch auctions ever since 1958, with around 900,000 gladioli stems/year in sales (Malter, 1995). Gladiolus corm production in the

Netherlands is 637 hectares while floral spike production was 153 hectares in 2018 (BDK 2019).

In China, gladiolus production covered 3300 hectares in 2014 which made it the second-most grown geophyte in China after lily (CoHort Consulting, 2018). In the United States, the total gladioli spike production is 61 million stems (USDA-ARS 2019) with a 2018 wholesale farmgate value of cut flower gladiolus of $20M (USDA-ARS, 2019).

1.4 Pests and diseases

There are several pest and diseases infecting gladiolus. Among the insects and other pests are mites, Rhizoglyphus rhizophagus (Bald & Jefferson 1952); thrips, Taeniothrips simplex

(Moulton & Steinweden 1931); wireworms, Agriotes sp. (McCulloch 1941) and cotton bollworm,

Helicoverpa armigera (Howard 1897). While nematodes infecting gladiolus include Meloidogyne incognita (Khanna & Chandel 1997). Diseases infecting gladiolus include bacteria, fungi,

Mycoplasma-like organisms (MLOs) and viruses. Bacterial diseases of gladiolus are:

Burkholderia gladioli pv. Pseudomonas gladiola (Severini 1913) and Xanthomonas campestris

ps. gummisudans pv. Bacterium gummisudans (McCuixoca 1924). A fungal disease, Fusarium oxysporum (Massey 1926), infects the corms and is the most destructive gladiolus disease (Dole

& Wilkins 2005). Other fungal diseases are caused by Alternaria brassicicola (Wu, He and et al

2020), Botrytis gladiolorum (Timmermans 1941), Curvularia lunata (Parmelee 1954),

Curvularia trifolii (Parmelee 1956), Penicillium gladioli (Machacek 1927), Rhizoctonia solani

(Creager 1945), Septoria gladioli (Massey 1916), Sclerotium rolfsii (Carpenter & Gammon

1955), Stemphylium botryosum (Nelson and Mooar 1949) and Stromatinia gladioli (Jeffers 1940).

Gladiolus are affected by a range of viruses and none of these viruses are host-specific to gladiolus species or Iridaceae, rather all these viruses infect a wide range of hosts (Stein 1995).

Common viruses infecting gladiolus are: Cucumber mosaic cucumovirus (CMV) (Wade 1948),

Bean yellow mosaic potyvirus (BYMV) (Brierleyl 1962), Tobacco ringspot (TRSV) (Smith &

Brierley 1955), Tomato ringspot (ToRSV) (Bozarth & Corbett 1958), Tomato black ring virus

(TBRV) (Kaminska 1978), Arabis mosaic virus (ArMV) (Bellardi & Marani 1984), Strawberry latent ringspot virus (SLRV) (Bellardi & Pisi 1983), Tobacco rattle tobravirus (TRV) (Cremer &

Schenk 1967), Tomato spotted wilt tospovirus (TSWV) (Lee, Francki & Hutta 1979).

Mycoplasma-like organisms (MLOs) also infect gladiolus (van Slogteren 1974). These diseases can cause destructive damage in gladiolus such as the MLOs that infect the corm phloem, leading to phloem degeneration and often to phloem necrosis (Bos 1995).

1.5 Gladiolus breeding

Gladiolus communis, G. italicus and G. byzantius were the first species introduced for gardening in England prior to 1730 (Wilfret 1980). The introduction of gladiolus species in

Europe piqued the interest of flower breeders. The first commercialized gladiolus hybrids were created by James Colville in Colville’s nursery in Chelsea, England, by crossing G. cardinalis to

G. tristis to create G. ×colvillei in 1823 (Beal and Hottes 1916). In the commercial market there are thousands of gladiolus cultivars in two major types. The first type is the primary cultivars used for cut flowers and gardening which are characteristically tall (heights reaching up to two meters) with long floral spikes, large-sized and a high number of multiple florets; all can be traced back to their ancestral hybrid parents G. ×gandevensis, G. ×lemoinei, G. ×childsii, and G.

×nanceianus (Okubo & Sochacki 2012). The other type of gladiolus is limited to use in gardens which are characterized by being shorter in height (reaching up to one meter), with multiple spikes, smaller floret sizes with “hooded” florets with the upper tepal being curved like a covering hood; their ancestral parent is G. dalenii (=G. primulinus) (Okubo & Sochacki 2012).

The Gladiolus genus has a base chromosome number of x = 15 in all species (Bamford

1935). Thus, diploid species have a total of 30 chromosomes (2n = 2x = 30), such as G. murielae

(Ohri 1985), while polyploids exist as: triploids (2n = 3x = 45), e.g. G. saundersii (Bamford

1935); tetraploids (2n = 4x = 60), i.e. G. bellus (Glodblatt 1993) and dodecaploids (2n = 12x =

176) in G. communis (Ohri 1985). It should be noted that the species native to Eurasia always exist in polyploids forms while the African species vary from diploids to polyploids (Goldblatt

1996). The current gladiolus hybrids in the market are mostly polyploids, primarily tetraploids with 60 chromosomes (2n = 4x = 60) (Bamford, 1935; Saito and Kusakari, 1972 and Ohri, 1985).

Singh (2014) addressed the breeding goals for gladiolus, such as overall improved growth, fragrance, flower shape, flower color, increased number of florets, increased vase life, increased production of corms / cormels and disease resistance. Some of these breeding objectives have been achieved. The interspecific hybrid, G. ×gandevensis, resulted in the creation of many significant cultivars which impacted breeding and development of subsequent cultivars such as ‘Picardy’ in 1932, a cultivar blend of G. ×gandevensis and G. primulinus (=G. dalenii).

This cultivar, bred by Palmer, was the first to introduce the cut flower in bud stage and

postharvest shipping capabilities to the cut flower market (Wilfret 1980). ‘Lucky Star’ was the first fragrant hybrid gladiolus, bred by Joan Wright in 1955, using G. murielae as the source of fragrance (Anderton and Park 1989). Additionally, G. callianthus has been used as a source of fragrance (Rao and Janakiram 1992). Another example is ‘Kundred’s Glory’, the first ruffled shape floret gladiolus, bred by A. E. Kundred in 1922 (Anderton and Park 1989). Breeding for disease resistance to Fusarium oxysporum has also commenced (Negi 1991).

1.6 Gladiolus genetics

Radiation was implemented to create mutants and test the variation that could be acquired through mutation breeding. One study resulted in flower color changes into lighter and darker flower colors (Kasumi 2001).

Random Amplified Polymorphic DNA (RAPDs) have been used to establish the genetic relationship between Gladiolus species hybrids and mutants through (Takatsu, 2001; Pathani and

Misra 2001; Wang et al 2009). Amplified Fragment Length Polymorphisms (AFLPs) were used to detect differences in radiation-induced mutants (Liu et al 2009). Inter-Simple Sequence

Repeats (ISSRs) were used to test differences generated in gladiolus hybrids treated with Ethyl methanesulfonate (EMS) (Gong et al 2010) as well as for genetic relatedness among winter-hardy genotypes (Anderson, et al., 2012).

Genetic transformation of gladiolus has been done using particle gun bombardment in callus (Kamo et al 1995a) and cormels (Kamo et al 1995b) as well as Agrobacterium-mediated transformation of callus (Babu and Chawla 2000). Transformation enabled use of such techniques to study the function genes in gladiolus and to produce transgenic gladiolus with resistance to

CMV (Kamo et al 2010), Fusarium resistance (Lakshman et al 2012) and resistance to BYMV

(Kamo et al 2005). Expansin GgEXPA1 (Azeeze et al 2010) and Plasma membrane Intrinsic

Proteins Ghpip1;1 (Hong-mei and Sheng-gen et al 2013) were the first genes identified in gladiolus. Cytochrome 450 gene GgCyP1 and the defender against cell death gene GgDAD1

(Dwivedi 2016), Ethylene receptor genes GgERS1a and GgERS1b (Ezura 2006) and ABSCICIC

ACID NSENSTIVE 5 gene GhABI5 (Wu et al 2015) have also been reported.

1.7 Gladiolus and the use of plant growth regulators

Plant growth regulators (PGRs) are bioactive chemicals that modulate plant growth during the life cycle as growth promoters, such as cytokinins, gibberellins, the inhibitory effect with ethylene and abscisic acid or both, growth promoters and signaling such as auxin (Taiz and

Zeiger 2010). They consist of naturally biosynthesized compounds in plants and they can also be artificially synthesized. The biosynthesized PGRs are plant hormones such as ethylene, zeatin, indole-3-acetic acid (IAA) and gibberellic acid (GA). The artificially synthesized PGRs are 1-

Naphthaleneacetic acid (NAA), Indole-3-butyric acid (IBA), 6-Benzyladenine (BA) and 2,4-

Dichlorophenoxyacetic acid (2, 4-D) and the whole class of plant growth retardants.

Application of PGRs are commonly used in horticulture. Auxin induces root formation in cuttings of tomato (Solanum esculentum) (Peer, 2019) and many other herbaceous floricultural crops and reduces pre-harvest drop in Citrus (Amiri, 2012). Spraying gibberellic acid prolongs postharvest storage of fruits such as (Diospyros kaki) persimmons (Ben-Arie, 1985), induces parthenocarpy in citrus (Randhawa, 1964), and increase fruit size in parthenocarpic (Malus sylvestris) crap apple (Bukovac & Nakagawa, 1967). Spraying Promalin® (a mixture of GA4+7 and BA) on apple (Malus Domestica) can increase fruit size, reduce fruit russeting and improve overall fruit ripening (Burak & Büyükyilmaz, 1997). Ethylene sprays induce flowering in

pineapple (Ananas comosus) (Cooke & Randall, 1968) and in juvenile mango plants (Mangifera indica) (Randhawa, 1974). Abscisic acid is used to enhance performance of potted plants and their postharvest survival by holding in water with the closure of stomata during long transportation periods of potted plants, such as chrysanthemum (Craig & Olrich 2011). Jasmonic acid is used to induce resistance to pests in grapes (Omar, 2000) and Salicylic acid applications induced resistance to Fusarium in tomato (Mandal, 2009).

In addition to the classic application of PGRs through spraying or drenching, transgenic crops carrying plant hormone(s) is an effective method of combining the knowledge of PGRs and biotechnology. An example of such is PSARK::IPT which can trigger the expression of the cytokinin biosynthesizing enzyme, Adenosine phosphate-isopentenyl transferase (IPT), under drought conditions to delay programmed cell death and survive drought conditions for a period of

7 – 14 days in tobacco (Gan and Amasino, 1995; Blumwald, 2010), rice (Blumwald, 2011) and peanuts (Qin, 2011) until irrigation can be applied.

Plant growth retardants are used to alter the growth of plants and gibberellin inhibitors are contained within a wide range of plant growth retardant products. Their primary use in commercial plant production is to reduce plant height through decreased cell elongation and cell division (Rademacher, 2016). Examples of plant growth retardants are Ancymidol, chlormequat chloride, Daminozide, flurprimidol, paclobutrazol and uniconazole (Whipker and Evans 2012). In woody ornamental plants, the commercial application of such growth retardants is used to restrict plant height and establish a uniform height, such as in the drench application of uniconazole to cause an inhibition of growth in woody landscape species (Warren, 1991).

In geophytes, the application of gibberellin inhibitors restricts plant height and establishes a uniform height for potted production geophytes in Hippeastrum (Miller et al., 2012) and Lilium

(Miller et al., 2002). Also, ancymidol and paclobutrazol are reported to delay flowering in Tulipa

(McDaniel, 1990) and Lilium (Bailey and Miller, 1989), while in other geophytic species different concentration can inhibit flowering, as reported in different Watsonia species (Ascough et al.,

2006). In gladiolus, exogenous applications of gibberellins can hasten flowering (Tonecki, 1980;

Sudhakar and Kumar, 2012), while gibberellin inhibitors cause reductions in plant height (Ahmad et al., 2014) and decrease the number of cormels formed (Steinitz and Lilien-Kipnis, 1989). Foliar application of growth using plant bioregulators have proved to be effective in gladiolus, plant bioregulators are unique agrochemical in that they must be absorbed by plant tissue and transported to a reaction site before the desired can be induced (Bukovac, 1997), polyamines putrescine and spermine increased plant height, spike height and diameter, the number of florets, diameter and weight of 1st flowering floret, corm diameter and weight, cormel production and their relative weights (Sajjad et al., 2015). Thus, exogenous applications of plant growth regulators can optimize the desired phenotype of ornamental plants in crop production to meet the market demands and serve as alternative methods to overcome some difficult to achieve traits in breeding.

1.8 Flowering models and genes

The transition from vegetative to reproductive growth is governed by flowering genes in which expression is influenced by factors such as vernalization, photoperiod, gibberellins, the autonomous pathway and temperature (Srikanth and Schmid 2011). This transition involves physiological and gene expression changes in plants. In Arabidopsis, several flowering genes were discovered that are involved in flowering and act as floral integrators. Some of these flowering genes are FT, SOC1, CO, VRN1, PPD, FCA, FLD, and FLK (Simpson & Dean 2002;

Simpson, 2004). Environmental stimuli modify the expression of these genes. However, the floral integrators gene is also reulated by repressor genes such as FLC, FRI, FLX, VRN2, and SVP

(Simpson & Dean 2002; Dean et al 2002; Ahn et al 2007).

FLC and FLC-like are floral repressor genes found in many dicotyledon plants (Peacock,

2001; Chalfun-Junior, 2012; Singh, 2016). The presence of FLC protein is regulated by temperature changes throughout the year, both in annuals and perennials. In the summer, FLC expression is upregulated through FRIGIDA (FRI) by binding the FLC promoter through the

DNA-binding protein SUPPRESSOR OF FRIGIDA4 SUF4 (Choi et al. 2011). FRI expression is upregulated by the EXPRESSOR OF FLOWERING LOCUS C (FLX); both SUF4 and FLX are in the FRI-specific pathway (Michaels et al. 2013). In contrast, in the winter, FLC is down-regulated through a process of vernalization as prolonged exposure of low temperature in winter in the meristem gradually reduces the expression of FLC (Michaels and Amasino 1999). In addition to the vernalization pathway, the autonomous pathway reduces the expression of FLC both in the meristem and leaves (Michaels and Amasino 1999). Gradual reduction of FLC allows the

FLOWERING LOCUS T (FT) to be expressed in the leaves and transported through phloem to reach to the meristematic tissue to stimulate the MADS box genes, which thereby induces flowering in Arabidopsis (Amasino 2005).

Vernalization is defined as the acceleration of flowering by a long period of cold, moist temperatures (Le Corre, 2006). In wheat and barley, the flowering pathway is regulated by photoperiod, vernalization and the circadian clock (Turner 2013). VRN2 acts a flowering inhibitor and, only through vernalization, is the expression of it downregulated. Then, the floral integrator leads to flowering in winter wheat while spring wheat does not require vernalization as the VRN2 gene is nonfunctional. Even though vernalization will speed up flowering in spring wheat, it acts as a facultative stimulus (Dubcovsky 2004). In contrast, maize and rice rely on plant age to build

up sufficient energy requirements in order to transition the vegetative plant into flowering through epigenetic action of miR172 (Helliwell 2011).

In monocotyledon geophytes, the flowering process is poorly understood. Factors of plant growth influencing flowering in commercial geophytes are well known (Ehrich, 2013). On the other hand, the genetic pathway is still in the early stages of discovery and characterization, in comparison to the Arabidopsis model which can be applied to temperate dicotyledon plants

(Fadón et al 2015). Only few flowering genes have been discovered in geophytes (Kamenetsky,

Zaccai and Flaishman, 2012), such as FT-like in Allium cepa (Taylor 2009; Taylor et al 2010), FT in Narcissus (Noy-Porat 2009), NLF in Narcissus (Noy-Porat 2010) and LFY in Allium sativum

(Rotem et al 2007; Rotem et al 2011), and LFY in Lilium (Wang et al 2008). Recently, many flowering genes have been discovered in Lilium ×formolongi, including FT, CO-like, AP2, GA1 and SOC1 (Jia et al 2017) while others, such as VER1 and VER2, have been proposed in Lilium

×formolongi (Zlesak and Anderson, 2009). The discoveries of flowering genes in geophytes serve as valuable resources to draw the model pathway of flowering geophytes.

Gladiolus is a monocotyledon with both summer and winter flowering species, FLC was not identified. It’s been hypothesized that there is no FLC gene in any monocotyledon species, until recently reported that FLC homologue were discovered in some cereal such as wheat (Sun,

UPSTREAM OF FLOWERING LOCUS C (UFC) and DOWNSTREAM OF FLOWERING

LOCUS C (DFC) in Arabidopsis (Finnegan et al 2004). UFC gene expression is repressed by vernalization, independent of FLC repression by vernalization (Finnegan et al 2004). Thus, both

FLC and UFC are repressed by vernalization, yet both are not dependent on each other for

expression; the suppression is through chromatin modification in an epigenetic manner (Finnegan et al 2004). The VRN1 gene is expressed with vernalization and acts as a floral integrator whereas the UFC gene is repressed and required by VRN1 expression dependently (Sheldon et al 2009).

1.9 Literature Cited

Ahmad, I., B.E. Whipker, and J.M. Dole. (2014). Flurprimidol preplant corm soaks or drenches

affect potted gladiolus production. Acta Hort. 1104:103-108.

Amiri, N.A., Kangarshahi, A.A., & Arzani, K. (2012). Reducing of citrus losses by spraying of

synthetic auxins. International Journal of Agriculture and Crop Sciences, 4(22), 1720-1724.

Anderson, N.O., J. Frick, A. Younis, and C. Currey. (2012). Heritability of cold tolerance (winter

hardiness) in Gladiolus xgrandiflorus. InTech Plant Breeding, ISBN 978-953-307-932-5.

Anderton, E.W., & Park, R. (1989). Growing gladioli. Timber Press.

Arora, A., Watanabe, S., Ma, B., Takada, K., & Ezura, H. (2006). A novel ethylene receptor

homolog gene isolated from ethylene-insensitive flowers of gladiolus (Gladiolus grandiflora

hort.). Biochemical and biophysical research communications, 351(3), 739-744.

Ascough, G.D., N.P. Mtshali, D.I. Thompson, N.O. Anderson, J.E. Erwin, and J. Van Staden.

2006. Watsonia. Floriculture, ornamental and plant biotechnology 4:559-605.

Azeez, A., Sane, A.P., Tripathi, S. K., Bhatnagar, D., & Nath, P. (2010). The gladiolus GgEXPA1

is a GA-responsive alpha-expansin gene expressed ubiquitously during expansion of all floral

tissues and leaves but repressed during organ senescence. Postharvest biology and

technology, 58(1), 48-56.

Babu, P., & Chawla, H.S. (2000). In vitro regeneration and Agrobacterium mediated

transformation in gladiolus. The Journal of Horticultural Science and Biotechnology, 75(4),

400-404.

Bald, J.G., & Jefferson, R.N. (1952). Injury to gladioli associated with the root mite,

Rhizoglyphus rhizophagus. Plant Dis. Reptr, 36, 435-437.

Bailey, D.A. and W.B. Miller. 1989. Whole-plant response of Easter lilies to ancymidol and

uniconazole. J. Amer. Soc. Hort. Sci. 144(3):393-396.

Bamford, R., 1935. The chromosome number in Gladiolus. Jour. Agr. Res, 51, pp.945-950.

Barreto, H. G., Lazzari, F., Ságio, S. A., Chalfun-Junior, A., Paiva, L. V., & Benedito, V. A.

(2012). In silico and quantitative analyses of the putative FLC-like homologue in coffee

(Coffea arabica L.). Plant Molecular Biology Reporter, 30(1), 29-35.

BDK. (2019). Voorlopige statistiek Gladiolus 2019. Available from: https://www.bkd.eu/wp-

content/uploads/2019/09/voorlopige-statistiek-gladiolus-2019-versie-3-18-09-2019-

website.pdf

Beal, A.C, Hottes, A.C. (1916) Gladiolus studies. New York state college of agriculture, New

York

Bellardi, M.G., & Pisi, A. (1983). Strawberry latent ringspot virus (SLRV) isolated from

gladiolus (Gladiolus sp.). 4th Int. Cong. PI. Path., Melbourne Australia. Australian Academy

of Science.

Bellardi, M.G., & Pisi, A. (1985). Survey of gladiolus viruses in Italy. Rivista di

ortoflorofrutticoltura italiana, 133-144.

Ben-Arie, R., Bazak, H., & Blumenfeld, A. (1985). Gibberellin delays harvest and prolongs

storage life of persimmon fruits. In V International Symposium on Growth Regulators in

Fruit Production 179 (pp. 807-814).

Bos, L. (1995). The economic importance of ornamentals. Virus and virus-like diseases of bulb

and flower crops. Wiley, New York, 85-98.

Bozarth, R.F., & Corbett, M.K. (1958). Tomato ringspot virus associated with stunt or stub head

disease of Gladiolus in Florida. Plant Disease Reporter, 42(2).

Brierleyl, P. (1962). Vol. 46, No. 7—Plant diseases reporter --July 15, 1962 505 Transmission of

some gladiolus viruses on tools in harvesting flowers and corms. The Plant Disease

Reporter, 46, 505.

Bukovac, M. J., & Nakagawa, S. (1967). Comparative potency of gibberellins in inducing

parthenocarpic fruit growth in Malus sylvestris mill. Experientia, 23(10), 865-865.

Burak, M., & Büyükyilmaz, M. (1997). Effect of Promalin on fruit shape and quality of starking

Delicious apple cultivar. In VIII International Symposium on Plant Bioregulation in Fruit

Production 463 (pp. 365-370).

Carpenter, T.R., & Gammon, E.T. (1955). Methyl bromide gas treatments of dormant gladiolus

corms and of sclerotia of Sclerotium rolfsii. Phytopathology, 45(9), 520-521.

CBS. (2018). Flower bulb cultivation continues to flourish. Available from:

https://www.cbs.nl/en-gb/news/2018/12/flower-bulb-cultivation-continues-to-flourish

Chacko, E.K., Kohli, R.R., Swamy, R.D., & Randhawa, G.S. (1974). Effect of (2‐Chloroethyl)

phosphonic Acid on Flower Induction in Juvenile Mango (Mangifera indica)

Seedlings. Physiologia Plantarum, 32(2), 188-190.

Choi, K., Kim, J., Hwang, H. J., Kim, S., Park, C., Kim, S. Y., & Lee, I. (2011). The FRIGIDA

complex activates transcription of FLC, a strong flowering repressor in Arabidopsis, by

recruiting chromatin modification factors. The Plant Cell, 23(1), 289-303.

Cohat, J. (1993). Gladiolus. W: The physiology of flower bulbs. Red. De Hertogh, A. & Le Nard,

M. Elsevier, Amsterdam, p. 297-320.

CoHort Consulting. (2018). Overview of the floricultural sector in China, 2018.

Cooke, A.R., & Randall, D.I. (1968). 2-Haloethanephosphonic acids as ethylene releasing agents

for the induction of flowering in pineapples. Nature, 218(5145), 974-975.

Craig, D., Olrich, M., & Runkle, E. (2011). Using ABA to Reduce. Greenhouse Grower. P. 31-33

Creager, D.B. (1945). Rhizoctonia neck rot of gladiolus. Phytopathology, 35, 230-232.

Cremer, M.C., & Schenk, P.K. (1967). Notched leaf in Gladiolus spp., caused by viruses of the

tobacco rattle virus group. Netherlands Journal of Plant Pathology, 73(2), 33-48.

Ding, L., Kim, S. Y., & Michaels, S. D. (2013). FLOWERING LOCUS C EXPRESSOR family

proteins regulate FLOWERING LOCUS C expression in both winter-annual and rapid-cycling

Arabidopsis. Plant physiology, 163(1), 243-252.

Dole, J.M., & Harold, W.F. (2005). Floriculture. Principles and Species. Prentice-Hall Inc.

Dwivedi, S.K., Arora, A., Singh, V.P., Sairam, R., & Bhattacharya, R.C. (2016). Effect of sodium

nitroprusside on differential activity of antioxidants and expression of SAGs in relation to

vase life of gladiolus cut flowers. Scientia Horticulturae, 210, 158-165.

Ehrich, L. (2013). Flowering in South African Iridaceae. (eds) Ramawat, K. G., & Merillon, J.

M. Bulbous Plants: Biotechnology. P. 248-269.

Fadón, E., Herrero, M., & Rodrigo, J. (2015). Flower bud dormancy in Prunus species.

In Advances in Plant Dormancy (pp. 123-135). Springer, Cham.

Finnegan, E. J., Sheldon, C. C., Jardinaud, F., Peacock, W. J., & Dennis, E. S. (2004). A cluster

of Arabidopsis genes with a coordinate response to an environmental stimulus. Current

Biology, 14(10), 911-916.

Gan, S., & Amasino, R. M. (1995). Inhibition of leaf senescence by autoregulated production of

cytokinin. Science, 270 (5244), 1986-1988.

Goldblatt, P., Takei, M., & Razzaq, Z.A. (1993). Chromosome cytology in tropical African

gladiolus (Iridaceae). Annals of the Missouri Botanical Garden, 461-470.

Goldblatt, P. (1996). Gladiolus in tropical Africa. Systematics, biology & evolution. Timber

Press, Inc..

Goldblatt, P., & Manning, J.C. (2008). The Iris family: natural history & classification. Timber

Press.

Goldblatt, P., Manning, J.C., von Blittersdorff, R., & Weber, O. (2014). New species of Gladiolus

L. and Moraea Mill. (Iridaceae) from Tanzania and Mozambique. Kew Bulletin, 69(2), 9496.

Goldblatt, P., Manning, J.C., Wightman, N., & Bytebier, B. (2019). Gladiolus virgineus, a new

species from Zambia and notes on G. ferrugineus in South Africa (Iridaceae:

Crocoideae). South African Journal of Botany, 123, 293-295.

Gong, S., Fu, H., & Wang, J. (2010). ISSR analysis of M1 generation of Gladiolus hybridus Hort

treated by EMS. Journal of Northeast Agricultural University. 17(2), 22-26.

Guan, L., Tayengwa, R., Cheng, Z.M., Peer, W.A., Murphy, A.S., & Zhao, M. (2019). Auxin

regulates adventitious root formation in tomato cuttings. BMC plant biology, 19(1), 435.

Halevy, A.H. (1986). The induction of contractile roots in Gladiolus grandiflorus. Planta, 167(1),

94-100.

Hertogh, A.D. (1996). Holland bulb forcer's guide (No. Ed. 5). International Flower Bulb Centre

and Dutch Bulb Exporters Association.

Howard, L.O. (1897). Insects affecting the cotton plant (No. 47). US Dept. of Agriculture.

International flower bulb center. (2010). Gladiolus as cut flowers Guidelines for cut flower

production. The International Flower Bulb Centre. Available from: https://www.vws-

flowerbulbs.nl/media/org/15c3e603db03a63d21e9e2cdbf4ea550.pdf

Jeffers, R. (1940). Gladiolus diseases. II. Gladiolus dry rot (Sclerotium gladioli). III. Gladiolus

hard rot (Septoria gladioli). Gladiolus diseases. II. Gladiolus dry rot (Sclerotium gladioli). III.

Gladiolus hard rot (Septoria gladioli).

Kamenetsky R, Zaccai M, Flaishman MA (2012) Florogenesis. In: Kamenetsky R, Okubo H

(eds) Ornamental geophytes: from basic science to sustainable production. CRC press,

Florida, pp 197-232

Kaminska, M., & Wos, M. (1978). Some properties of isolates of Tomato black ring

virus. Zeszyty Probl Post Nauk Roln, 214, 109-17.

Kamo, K., Blowers, A., Smith, F., Van Eck, J., & Lawson, R. (1995a). Stable transformation of

Gladiolus using suspension cells and callus. Journal of the American Society for Horticultural

Science, 120(2), 347-352.

Kamo, K., Blowers, A., Smith, F., & Van Eck, J. (1995b). Stable transformation of Gladiolus by

particle gun bombardment of cormels. Plant Science, 110(1), 105-111.

Kamo, K., Gera, A., Cohen, J., Hammond, J., Blowers, A., Smith, F., & Van Eck, J. (2005).

Transgenic Gladiolus plants transformed with the bean yellow mosaic virus coat-protein gene

in either sense or antisense orientation. Plant cell reports, 23(9), 654-663.

Kamo, K., Jordan, R., Guaragna, M.A., Hsu, H.T., & Ueng, P. (2010). Resistance to Cucumber

mosaic virus in Gladiolus plants transformed with either a defective replicase or coat protein

subgroup II gene from Cucumber mosaic virus. Plant cell reports, 29(7), 695-704.

Kasumi, M. (2001). Studies on induction of flower color mutants in gladiolus (Gladiolus x

grandiflora Hort.) by gamma irradiation and tissue culture. Bulletin of the Plant

Biotechnology Institute Ibaraki Agricultural Center (Japan).

Khanna, A.S., & Chandel, S.S. (1997). Damage potential of Meloidogyne incognita in gladiolus

cv Sylvia. Indian Journal of Nematology, 27(2), 253-254.

Kumar, G., Arya, P., Gupta, K., Randhawa, V., Acharya, V., & Singh, A. K. (2016). Comparative

phylogenetic analysis and transcriptional profiling of MADS-box gene family identified

DAM and FLC-like genes in apple (Malus x domestica). Scientific reports, 6(1), 1-13.

Lakshman, D.K., Pandey, R., Kamo, K., Bauchan, G., & Mitra, A. (2012). Genetic transformation

of Fusarium oxysporum f. sp. gladioli with Agrobacterium to study pathogenesis in

Gladiolus. European journal of plant pathology, 133(3), 729-738.

Lee, J. H., Yoo, S. J., Park, S. H., Hwang, I., Lee, J. S., & Ahn, J. H. (2007). Role of SVP in the

control of flowering time by ambient temperature in Arabidopsis. Genes &

development, 21(4), 397-402.

Lee, T.C., Francki, R.I.B. and Hatta, T. (1979). A serious disease of gladiolus in Australia caused

by tomato spotted wilt virus. Plant Dis. Rep. 63:343-348. 20

Levy, Y. Y., Mesnage, S., Mylne, J. S., Gendall, A. R., & Dean, C. (2002). Multiple roles of

Arabidopsis VRN1 in vernalization and flowering time control. Science, 297(5579), 243-246.

Lewis, G.J., Obermeyer, A.A., & Barnard, T.T. (1972). A revision of the South African species of

Gladiolus. JS Afr. Bot. Suppl, 10, 1-316.

Li, Y. F., Zhang, M. F., Zhang, M., & Jia, G. X. (2017). Analysis of global gene expression

profiles during the flowering initiation process of Lilium × formolongi. Plant molecular

biology, 94(4-5), 361-379.

Lin, Y.F., Li, H.M., Ding, Y.L., Huang, X.M., Xian, X.J., & He, S.G. (2013). Cloning and

expression analysis of a plasma membrane aquaporin gene GhPIP1,'I in cut gladiolus

flowers. Acta Horticulturae Sinica, 1.

Liu, C., Zhang, X., Xu, Q., & Xian, F. (2009). AFLP analysis on dual color flower gladiolus

mutants induced by radiation. Acta Botanica Boreali-Occidentalia Sinica, 29(1), 69-74.

Malter, A.J. (1995). The economic importance of ornamentals. Virus and virus-like diseases of

bulb and flower crops. Wiley, New York, 1-13.

Machacek, J.E. (1927). A Penicillium rot of Gladiolus.

Mandal, S., Mallick, N., & Mitra, A. (2009). Salicylic acid-induced resistance to Fusarium

oxysporum f. sp. lycopersici in tomato. Plant physiology and Biochemistry, 47(7), 642-649.

Massey, L.M. (1916). The hard rot disease of Gladiolus (Vol. 380). Cornell University.

Massey, L.M. (1926). Fusarium rot of gladiolus corms. Phytopathology, 16, 509-523.

McCuixoca, L. U. C. I. A. (1924). A bacterial blight of gladioli'. Journal of Agricultural

Research, 27, 225.

McCulloch, L. (1941). Gladiolus diseases and insects (No. 1860). US Department of Agriculture.

McDaniel, G.L. 1990. Postharvest height suppression of potted tulips with

paclobutrazol. HortScience 25(2):212-214.

Michaels, S.D. and Amasino, R.M. (1999) FLOWERING LOCUS C encodes a novel MADS

domain protein that acts as a repressor of flowering. Plant Cell 11, 949–956

Michaels, S. D., Himelblau, E., Kim, S. Y., Schomburg, F. M., & Amasino, R. M. (2005).

Integration of flowering signals in winter-annual Arabidopsis. Plant Physiology, 137(1), 149-

156.

Miller, C.T., C.M. Filios, and W.B. Miller. 2012. Effects of flurprimidol, paclobutrazol and

uniconazole soaks and drenches on amaryllis (Hippeastrum) growth and development. Acta

Hort. 1002:431-438.

Monge, A.V. (1981). Monejo de los cormos de gladiolos. Hojas divulgadoras-Spain, Ministerio

de Agricultura.

Monteagudo, A., Igartua, E., Contreras-Moreira, B., Gracia, M. P., Ramos, J., Karsai, I., & Casas,

A. M. (2019). Fine-tuning of the flowering time control in winter barley: the importance of

HvOS2 and HvVRN2 in non-inductive conditions. BMC plant biology, 19(1), 1-14.

Moulton, D., & Steinweden, J.B. (1931). A new Taeniothrips on gladiolus. The Canadian

Entomologist, 63(1), 20-21.

Negi, S.S., Raghava, S.P.S., Chacko, C.I., Rao, T.M., & Ramachandran, N. (1991). Breeding for

quality and resistance to fusarial wilt in gladiolus. In Horticulture—New Technologies and

Applications (pp. 21-25). Springer, Dordrecht.

Nelson, R., & Mooar, M. (1949). Life history of Stemphylium causing red leaf spot of gladiolus.

In Phytopathology (Vol. 39, No. 1, pp. 16-17). 3340 Pilot Knob Road, St Paul, MN 55121

USA: Amer Phytopathological Soc.

Noy-Porat, T., Flaishman, M. A., Eshel, A., Sandler-Ziv, D., & Kamenetsky, R. (2009).

Florogenesis of the Mediterranean geophyte Narcissus tazetta and temperature requirements

for flower initiation and differentiation. Scientia horticulturae, 120(1), 138-142.

Noy-Porat, T., Kamenetsky, R., Eshel, A., & Flaishman, M. A. (2010). Temporal and spatial

expression patterns of the LEAFY homologue NLF during florogenesis in Narcissus

tazetta. Plant Science, 178(2), 105-113.

Ohri, D., 1985. Cytogenetics of Garden Gladiolus. Cytologia, 50(2), pp.213-231.

Okubo, H., & Sochacki, D. (2012). Botanical and horticultural aspects of major ornamental

geophytes. Ornamental geophytes: from basic science to sustainable production. Boca Raton:

CRC Press, Taylor and Francis Group, 77-117.

Omer, A.D., Thaler, J.S., Granett, J., & Karban, R. (2000). Jasmonic acid induced resistance in

grapevines to a root and leaf feeder. Journal of economic entomology, 93(3), 840-845.

Parmelee, J.A. (1954). Curvularia on gladiolus in Canada. Plant Dis. Reptr, 38(7), 517.

Parmelee, J.A. (1956). The identification of the Curvularia parasite of

gladiolus. Mycologia, 48(4), 558-567.

Pate, J.S., & Dixon, K.W. (1982). Tuberous, cormous and bulbous plants. International Scholarly

Book Services Inc.

Pathania, N.S., & Misra, R.L. (2001). Characterization of gladiolus mutants using RAPD

markers. Journal of Ornamental Horticulture, 4(2), 65-68.

Peleg, Z., Reguera, M., Tumimbang, E., Walia, H., & Blumwald, E. (2011). Cytokinin‐mediated

source/sink modifications improve drought tolerance and increase grain yield in rice under

water‐stress. Plant Biotechnology Journal, 9(7), 747-758.

Peri, O. (2015). Bulbs of the eastern Mediterranean. Alpine Garden Society. UK

Pickering, H., & Patzelt, A. (2008). Field guide to the wild plants of Oman. Royal Botanic

Gardens.

Prasad, V.S.S., & Gupta, S.D. (2006). In vitro shoot regeneration of gladiolus in semi-solid agar

versus liquid cultures with support systems. Plant cell, tissue and organ culture, 87(3), 263-

271.

Qin, H., Gu, Q., Zhang, J., Sun, L., Kuppu, S., Zhang, Y., Burow, M., Payton, P., Blumwald, E.,

& Zhang, H. (2011). Regulated expression of an isopentenyltransferase gene (IPT) in peanut

significantly improves drought tolerance and increases yield under field conditions. Plant and

Cell Physiology, 52(11), 1904-1914.

Randhawa, G.S., Dhuria, H.S., & Nair, P.K.R. (1964). A note on gibberellin induced

parthenocarpy in citrus. Indian Journal of Horticulture, 21(2), 171-172.

Ranwala, A.P., G. Legnani, M. Reitmeier, B.B. Stewart, and W.B. Miller. 2002. Efficacy of plant

growth retardants as preplant bulb dips for height control in LA and oriental hybrid

lilies. HortTechnology, 12(3):426-431.

Rao, T.M. and Janakiram, T. (1992). Breeding for fragrance in gladiolus utilizing wild species.

Abst: National Seminar on Commercial Floriculture in India: Present & Potential, Bangalore.

P. 36.

Rees, A.R. (1992). Ornamental bulbs, corms and tubers. CAB international.

Rivero, R. M., Gimeno, J., Van Deynze, A., Walia, H., & Blumwald, E. (2010). Enhanced

cytokinin synthesis in tobacco plants expressing PSARK::IPT prevents the degradation of

photosynthetic protein complexes during drought. Plant and cell physiology, 51(11), 1929-

1941.

Rotem, N., Shemesh, E., Peretz, Y., Akad, F., Edelbaum, O., Rabinowitch, H. D., Sela, I., &

Kamenetsky, R. (2007). Reproductive development and phenotypic differences in garlic are

associated with expression and splicing of LEAFY homologue gaLFY. Journal of

Experimental Botany, 1133-1141.

Rotem, N., David-Schwartz, R., Peretz, Y., Sela, I., Rabinowitch, H. D., Flaishman, M., &

Kamenetsky, R. (2011). Flower development in garlic: the ups and downs of gaLFY

expression. Planta, 233(5), 1063-1072.

Roux, F., Touzet, P., Cuguen, J., & Le Corre, V. (2006). How to be early flowering: an

evolutionary perspective. Trends in plant science, 11(8), 375-381.

Ruelens, P., De Maagd, R. A., Proost, S., Theißen, G., Geuten, K., & Kaufmann, K. (2013).

FLOWERING LOCUS C in monocots and the tandem origin of angiosperm-specific MADS-

box genes. Nature Communications, 4(1), 1-8.

Saito, K. and Kusakari, K. (1972). Studies on the Occurrence of Polyploidy and its Contribution

to the Flower Plants Breeding.: IX. Cytological Observations on the Mechanism of Decreased 25

Fertility in the Summer-flowering Tetraploid Cultivars of Gladiolus grandiflorus

Hort. Japanese Journal of Breeding, 22(2), pp.75-82.

Sajjad, Y., Jaskani, M.J., Qasim, M., Akhtar, G., & Mehmood, A. (2015). Foliar Application of

Growth Bioregulators Influences Floral Traits, Corm-associated Traits and Chemical

Constituents in Gladiolus grandiflorus L. 원예과학기술지, 33(6), 812-819.

Seng, S., Wu, C., Wu, J., Zhong, X., He, J., & Yi, M. (2020). Silencing GhCOI1 in Gladiolus

hybridus increases susceptibility to Alternaria brassicicola and impairs inducible

defenses. Plant Cell, Tissue and Organ Culture (PCTOC), 140(1), 69-81.

Serek, M., Jones, R.B., & Reid, M.S. (1994). Role of ethylene in opening and senescence of

Gladiolus sp. flowers. Journal of the American Society for Horticultural Science, 119(5),

1014-1019.

Severini, G. (1913). Una bacteriosi dell'lxia maculate e del Gladiolus. Colvilli. Ann. Bot.

(Rome), 11, 413-424.

Sheasby, P. (2007). Bulbous plants of Turkey and Iran: (including the adjacent Greek Islands); a

photographic guide. Alpine Garden Society.

Sheldon, C. C., Jean Finnegan, E., James Peacock, W., & Dennis, E. S. (2009). Mechanisms of

gene repression by vernalization in Arabidopsis. The Plant Journal, 59(3), 488-498.

Simpson, G. G. (2004). The autonomous pathway: epigenetic and post-transcriptional gene

regulation in the control of Arabidopsis flowering time. Current opinion in plant

biology, 7(5), 570-574.

Simpson, G. G., & Dean, C. (2002). Arabidopsis, the Rosetta stone of flowering

time?. Science, 296(5566), 285-289.

Singh, A.K. (2006). Flower crops: cultivation and management. New India Publishing. P. 147-

166.

Singh, A.K. (2014). Breeding and biotechnology of flowers. New India Pub. Agency.

Smith, F.F., & Brierley, P. (1955). Aphid transmission of tobacco ringspot virus in

gladiolus. Plant Dis. Rep, 39, 35.

Stein, A. (1995). Gladiolus. Virus and virus-like diseases of bulb and flower crops. Wiley, New

York, 281-292.

Steinitz, B. and H. Lilien-Kipnis. 1989. Control of precocious gladiolus corm and cormel

formation in tissue culture. J. of plant physiology, 135(4):495-500.

Srikanth, A., & Schmid, M. (2011). Regulation of flowering time: all roads lead to

Rome. Cellular and molecular life sciences, 68(12), 2013-2037.

Sudhakar, M. and S.R. Kumar. 2012. Effect of growth regulators on growth, flowering and corm

production of gladiolus (Gladiolus grandiflorus L.) cv. White friendship. Ind. J. Plant

Sci, 1(2/3):133-136.

Tadege, M., Sheldon, C. C., Helliwell, C. A., Stoutjesdijk, P., Dennis, E. S., & Peacock, W. J.

(2001). Control of flowering time by FLC orthologues in Brassica napus. The Plant

Journal, 28(5), 545-553.

Taiz, L., & Zeiger, E. (2010). Plant physiology 5th Ed. Sunderland, MA: Sinauer Associates, 464.

Takatsu, Y., Miyamoto, M., Inoue, E., Yamada, T., Manabe, T., Kasumi, M., Hayashia, M.,

Sakumac, F., Marubashid, W., & Niwa, M. (2001). Interspecific hybridization among wild

Gladiolus species of southern Africa based on randomly amplified polymorphic DNA

markers. Scientia horticulturae, 91(3-4), 339-348.

Taylor, A. (2009). Functional genomics of photoperiodic bulb initiation in onion (Allium

cepa) (Doctoral dissertation, University of Warwick).

Taylor, A., Massiah, A. J., & Thomas, B. (2010). Conservation of Arabidopsis thaliana

photoperiodic flowering time genes in onion (Allium cepa L.). Plant and cell

physiology, 51(10), 1638-1647.

Timmermans, A.S. (1941). Het Botrytis-rot der Gladiolen, veroorzaakt door Botrytis gladiolorum

nov. spec. Med. Lab. v. Bloembollenonderzoek, Lisse, (67), 32.

Tomiozzo, R., Uhlmann, L.O., Becker, C.C., Schwab, N.T., Streck, N.A., & Balest, D.S. (2019).

How to produce gladiolus corms?. Ornamental Horticulture, 25(3), 299-306.

Tonecki, J. 1980. Effect of growth regulators on shoot apex differentiation and changes in sugars

and free amino acids in gladiolus (gladiolus × hortorum cv. acca laurentia). Acta Hort.

109:347-356.

Turner, A. S., Faure, S., Zhang, Y., & Laurie, D. A. (2013). The effect of day-neutral mutations

in barley and wheat on the interaction between photoperiod and vernalization. Theoretical

and applied genetics, 126(9), 2267-2277.

U.S. Department of Agriculture. National Agricultural Statistics Service 2019. Floriculture 2018.

USDA Economics, Statistics and Market Information System, Albert R. Mann Library,

Cornell University, Ithaca, New York. Available from: 28

https://downloads.usda.library.cornell.edu/usda-

esmis/files/0p0966899/rr1728124/76537c134/floran19.pdf van Slogteren, D.H.M. (1974). Mycoplasmas in bulb crops, particularly in gladiolus and

hyacinth. Antonie van Leeuwenhoek, 40(2), 314-314.

Wade, G.C. (1948) Gladiolus diseases. Tasmania J. Agric. 19:36-40

Wang, A. J., Tang, J. F., Zhao, X. Y., & Zhu, L. H. (2008). Isolation of LiLFY1 and its expression

in lily (Lilium longiflorum Thunb.). Agricultural Sciences in China, 7(9), 1077-1083.

Wang, J.G., Guo, Y., Che, D.D., Liu, S.K., & Yang, C.P. (2009). RAPD Analysis of M1

Generation of Gladiolus hybridus Hort Treated by EMS. Journal of Northeast Agricultural

University (English Edition), 16(1), 5-9.

Warren, S. L., Blazich, F. A., & Thetford, M. (1991). Whole-plant response of selected woody

landscape species to uniconazole. Journal of Environmental Horticulture, 9(3), 163-167.

Whipker, B.E. and M.R. Evans. 2012. Regulation of plant growth, p. 373-390. In: P.V.

Nelson (ed.). Greenhouse Operation and Management. 7th ed. Prentice Hall, Upper Saddle

River, NJ.

Wilfret, G.J. (1980). Gladiolus. In “Introduction to floriculture” (Larson RA ed.).

Wu, J., Seng, S., Sui, J., Vonapartis, E., Luo, X., Gong, B., Liu, C., Wu, C., Liu, C., Zhang, F.,

He, J., & Yi, M. (2015). Gladiolus hybridus ABSCISIC ACID INSENSITIVE 5 (GhABI5) is an

important transcription factor in ABA signaling that can enhance Gladiolus corm dormancy

and Arabidopsis seed dormancy. Frontiers in plant science, 6, 960.

Yan, L., Loukoianov, A., Blechl, A., Tranquilli, G., Ramakrishna, W., SanMiguel, P., Bennetzen,

J., Echenique, V. & Dubcovsky, J. (2004). The wheat VRN2 gene is a flowering repressor

down-regulated by vernalization. Science, 303(5664), 1640-1644.

Yasuda, I., & Yokoyama, N. (1955). Biochemical studies in breaking the rest period of gladiolus

corms. I. Journal of the Japanese Society for Horticultural Science, 24(3), 207-210.

Zlesak, D. and N.O. Anderson. (2009). Inheritance of non-obligate vernalization requirement for

flowering in Lilium formosanum Wallace. In: R. Kaminetsky (Ed.). Special Issue:

“Ornamental Geophytes”, Isr J of Plant Sci 57(4):315-327.

Zettler, F.W., FW, Z., & MM, A.E.N. (1977). Bean Yellow Mosaic Virus infections of Gladiolus

in Florida.

Zhao, T., Ni, Z., Dai, Y., Yao, Y., Nie, X., & Sun, Q. (2006). Characterization and expression of

42 MADS-box genes in wheat (Triticum aestivum L.). Molecular Genetics and

Genomics, 276(4), 334.

Zhu, Q. H., & Helliwell, C. A. (2011). Regulation of flowering time and floral patterning by

miR172. Journal of experimental botany, 62(2), 487-495.

2.0 Gladiolus breeding for rapid generation cycling and reduced dormancy

To be submitted to Euphytica

Jaser A. Aljaser1 and Neil O. Anderson2

Department of Horticultural Science, 1970 Folwell Avenue, University of Minnesota, St. Paul,

MN 55108 USA

1 Graduate Research Assistant 2 Professor; corresponding author email: [email protected] 31

Subject Category: Breeding, Cultivars, Rootstocks, and Germplasm Resources

Gladiolus breeding for rapid generation cycling and reduced dormancy

Additional index words. Gladiolus ×hybridus, flowering, geophyte, cycle 1 flowering, rapid generation cycling, dormancy.

2.1 Preface

The iconic gladiolus cut flower is and has been desirable in floral market internationally for centuries. This rise in demand requires the crop improvement through breeding strategies.

However, gladiolus is perennial geophyte with long juvenile period that can reach up to 5 years.

The long juvenile period, this forms a barrier that slows breeding. Here we show the results of constant breeding selection for two decades in order to reduce the juvenility period. Through the rapid generation cycling (RGC) method for early flowering gladiolus, we successfully converted the perennial juvenile gladiolus into an annualized perennial gladiolus with a range of 3 – 11 months from germination of seedling to flowering in the greenhouse. Further breeding through recurrent selection and inbreeding resulted in recurrent selections which have the significantly fastest flowering generation with 21.63 weeks faster than in RGC-2 than 31.60 weeks in RGC-1

(parents) and 39.00 weeks in RGC-3 (inbreds), while seed germination was significantly earliest in the parent generation in 2.65 weeks. In addition, some genotypes show unique ability of sprouting after completing a cycle without cold treatment to break the dormancy, which is more common hybrids than inbred genotypes. This reduced dormancy trait and the annualized 32

gladiolus genotypes can be implemented as tools to hasten the breeding process in gladiolus.

Also, the seed-propagated gladiolus could be marketable for potted production due to their petite aesthetic phenotype.

2.2 Introduction

Gladiolus ×hybridus Rodigas, also known by sword lily and commonly known as gladiolus, is the top cut flower in the floral markets. In USA, gladiolus ranked as the 4th cut flower with whole sales exceed $20 in 2018 (USDA 2019). Gladiolus is in the Iridaceae and native to South Africa and there are ~265 Gladiolus species, most of which are native to South

Africa and others across Africa, while approximate 10 species in Eurasia (Goldblatt and Manning

1998). Gladioli were introduced to Europe during the age of exploration around the 18th century, an example of these species are Gladiolus angustus, G. carenus, G. cardinalis and G. undulates, and all introduced species are from the Western Cape Province of South Africa (Beal and Hottes

1916). Since then more gladioli species been introduced and this drove for the interest in breeding for a variety of floral characteristics.

The history of first interspecific hybrid done by William Herbert in early 1806, showed that these hybrids appeared to be sterile and they were not commercialized (Randhawa and

Mukhopadhyay 1986). The first commercialized gladiolus hybrids were succeeded by James

Colville in Colville’s nursery at Chelsea, England; he crossed G. cardinalis to G. tristis resulting in G. ×colvillei in 1823 (Beal and Hottes 1916). This would propose the possibility that gladioli have the capacity to generate interspecific hybrids. With the flow of introducing new gladioli species to western Europe and interspecific hybridization, gladiolus became a popular plant to hybridize and competition to breed new colorful and vigorous gladiolus, these new hybrids were 33

estimated to be over 10,000 released cultivars (Sinha and Roy 2002) although many are no longer available (Rees 1992). These commercial gladiolus hybrids in the market are known as Gladiolus

×hybridus as summer hybrids gladiolus originate from several gladiolus species (Kamenetsky and

Okubo 2012). Furthermore, these interspecific hybrids are reported to have a basic chromosome number is x = 15 (Bamford 1941). Diploid gladioli (2n = 2x = 30) are common in the wild while commercial gladioli are typically polyploids, with a range of triploids to tetraploids.

Gladiolus breeding programs have breeding objectives for crop improvement including:

Fusarium resistance (Straathof et al. 1997), Stromatinia resistance (Eijk et al. 1989), virus resistance (Kamo et al. 2005), pest resistance (Zeier and Wright 1995), flower color & pattern changes (Kasumi 2001), cold tolerance for USDA zone 3-4 (Anderson et al. 2011), and flower fragrance. Gladiolus ‘Lucky Star’, bred by Joan Wright in 1966, is the first hybrid fragrant gladiolus (Anderton and Park 1989).

In commercial production, Gladiolus are planted as corms (a type of geophyte), which are compressed, underground stems. It is vegetatively propagated through cormel production, because seed-propagated gladiolus species would typically take multiple (3-5) years (or cycles) to flower, as the life cycle of wild gladiolus is starting from seed germination in the spring season into a seedling forming a small corm. The seedling continues in growth by usually having one or two leaves then the seedling undergoes the senescence process, resulting in deep dormancy for the corm in the fall until winter. During winter, the cold temperature breaks the dormancy by inhibiting the levels of abscisic acid and increasing gibberellin levels, resulting in corm sprouting in the spring (Wu et al. 2015). Vegetative leaf growth continues in spring and summer and usually senesce in the fall without flowering, thus repeating the cycle multiple times, ranging from 3 years to 5 years of the juvenile period until gladiolus corms finally reach an adequate level

of growth in order to flower in wild species (Anderson 2015) and as early as two years in some commercial cultivars (Le Nard et al. 1996) although it can commonly flower in the third year.

This long process of the juvenility period is also common in ornamental geophytes, such as

Tulipa with 4-6 years and Narcissus in 4-6 years (Fortanier 1973). The long juvenile period delays the speed of breeding in gladiolus.

A first-year flowering gladiolus was first reported in 1907 by Roemer (Roemer 1907).

Frederick Roemer intercrossed early flowering cultivars of G. ×gandevensis, G. ×lemoinei, G.

×childsii, and G. ×nanceianus resulting in G. ×praecox that bloomed in 1905 (Roemer 1907).

Although one older record could indicate that G. ×lemoine hybrids had a cycle 1 flowering pattern, G. ×lemoine is a hybrid between G. papilio and G. ×gandevensis. This was first reported in The Gladiolus, Its History, Species and Cultivation, by John Lewis Childs, the breeder who introduced G. ×childsi in the market (Childs, 1893). Moreover, this may suggest G. ×lemoine could be the source of the cycle 1 flowering pattern in G. ×praecox, as Roemer indicated it rarely happened that a seedling produced well-developed flowers in its first year of sowing (Roemer

1907). Additionally, there was a small frequency of cases of a cycle 1 gladiolus emerging in seedlings reported in the 1920s, originating from “quick-growing strains” (McLean, Clark, and

Fischer 1927). As early 1893 and to the 1930s, cycle 1 gladioli was reported in publications and were commercialized by the Burpee© Seed Company, as gladiolus ‘Fordhook hybrid’ in 1914, in which Burpee© bred their own cycle 1 cultivar using G. praecox (Burpee 1913).

Gladiolus praecox was bred to reduce the juvenile period. However, there are few of known species exhibiting cycle 1 flowering in their natural habitat, induced through horticultural practices or through breeding and selection. Freesia laxa, F. caryophyllacea, F. grandiflora, F. leichtlinii subsp. alba, F. leichtlinii subsp. leichtlinii, F. sparrmanii are all reported to flower in

their natural habitat in South Africa from seed in first year (Duncan 2010). Some of the Siberian irises, Iris sibirica and I. domestica, show a pattern of cycle 1 flowering (William Dougherty,

2017, personal communication; Anderson, 2019). It is also a known practice in iris breeding to stress the seedlings of beardless iris hybrids, e.g. I. siberica and I. spuria, with water and nutrients to force them to flower in the first year (Vaughn 2015). Crocosmia aurea and C. pottsii usually bloom from seed in the second year, However, if heat and light are “well provided”, both species and their hybrids flower from seed in the first year (Goldblatt, et al. 2004). Tigridia pavonia is a well-known geophyte flower species in which seedlings sometimes flower in the first year and is commercially sold in seeds packets and well as corms (De Hertogh and Le Nard 1993;

Kamenetsky and Okubo 2012). Lilium formosanum is the only known lily species which is able to flower from seed in the first year and much breeding research is being done in University of

Minnesota lily breeding program to annualize it as a seed-propagated lily. As an interspecific hybrid between L. formosanum x L. longiflorum, L. ×formolongi, also flowers in <120 days from sowing (Anderson et al. 2009).

Due to its economic importance, the University of Minnesota flower breeding program is breeding gladiolus for cold tolerance in USDA zone 3-4 (Anderson et al. 2011), with reduced generation cycling (Anderson et al. 2015; Anderson, 2019), as well as new flower colors and patterns (Figure 1). Breeding dwarf gladiolus for potted plant production and breeding for seed- propagated gladiolus are also new traits for this crop. The concept of rapid generation cycling

(RGC) is the ability to flower from seed in less than a year with a reduced juvenility and/or dormancy period (Anderson, 2019). The gladiolus breeding program at the University of

Minnesota started by selecting early germinating species and hybrids as well as crossing G.

×hybridus ‘Atom’ to G. primulinus ‘Carolina Primrose’, followed by selecting and crossing progeny (04GL generation: gladiolus crosses and selection done in year 2004 ) with G. ×hybridus 36

(‘Great Lakes’, ‘Beatrice’ and seedling no. 98-29), resulting in 06GL populations (gladiolus crosses and selection done in year 2006). In addition, G. ×hybridus 'Lady Lucille’ was crossed with ‘King’s Gold’. The 06GL populations were divided into two groups, with the first group planted in the field at the University of Minnesota Southern Research and Outreach Center in

Waseca, MN and the second group was planted in the St. Paul, MN greenhouses. Both groups were allowed to open pollinate, which formed the 13GL generation. The 13GL generation

(gladiolus crosses and selection done in year 2006) was where the 1st generation cycle 1 (RGC-1) gladioli plants occurred in greenhouse (Anderson, unpublished data; cf. Anderson, 2019).

The objective of this study is to breed new gladiolus genotypes able to flower in their first year to produce a seed-propagated gladiolus for their overall use in potted production of gladiolus. The hypothesis tested is H0: The mean number of weeks of seed germination and number of weeks to reach flowering is equal across the parent, recurrent selection and inbred generations (µ1 = µ2 = µ3). HA: The mean number of weeks of seed germination and number of weeks to reach flowering is not equal across the parent, recurrent selection and inbred generations

(µ1 ≠ µ2 ≠ µ3).

2.3 Material and Methods

Plant material. Gladiolus genotypes used for breeding are, Gladiolus dalenii ‘Carolina Primrose’,

G. ×hybridus ‘Atom’, with additional commercial cultivars G. ×hybridus (‘Great Lakes’,

‘Beatrice’, ‘King’s Gold’ and 'Lady Lucille’), and seedling no. 98-29 obtained from Donald R.

Selinger of the Minnesota Gladiolus Society.

Plant breeding. The Rapid Generation Cycling – 1 (RGC-1) has 20 cycle 1 plants and they are used as the parents for crosses, these plants are selected to cross pollinate with each other resulting in 15GL (gladiolus crosses and selection done in year 2015) and 16GL (gladiolus crosses and selection done in year 2016) recurrent selection (RGC-2 recurrent selection) and inbred of RGC-1 labeled RGC-2 inbred. The 16GL generation cycle 1 and cycle 1.5 plants are selected to self-pollinate, forming the 17GL (gladiolus crosses and selection done in year 2017) and 18GL generations (gladiolus crosses and selection done in year 2018) (RGC-3) (Figure 2).

Efficient pollination requires re-pollination for two successive days, resulting in larger capsules

(fruit) which can have up to 150 seeds per capsule (Figure 3).

Method of selection. Selection of new genotypes is done through “tooth-picking” by placing a tooth pick for any germinated gladiolus seedling, these seedlings are differentiated by each week of germination represented by “G” e.g. germination in first week is G1 and second week is G2 to eight-week G8 (Anderson, 2019). Selection of flowering for cycle 1 is through selecting plants which can flower in the first year of seed germination.

Greenhouse growing conditions. Gladiolus seed wings are removed to enhance early seed germination as gladiolus wing delay germination (Griesbach 1972) and seeds are sown in 288 seed plugs PL-288-1.5 (T.O. Plastics, Clearwater, MN) with germination mix mediu , “BM2”

(Berger, Saint-Modeste. QC). Seed plug trays were placed in a mist greenhouse. After germination, seed plugs are placed on capillary mats where they’re grown in a long day photoperiod (0800 – 1600 HR supplied by 400-W high-pressure sodium lamps + 2200 to 0200

HR night interruption, >150 µmol m-2 sec-1) at a minimum setpoint of 18o C (day/night), 70-80% relative humidity, with irrigation accomplished using constant liquid feed (CLF) of 125 ppm N from water-soluble 20N–4.4P–16.6K (Scotts, Marysville, OH) and deionized water on weekends.

After reaching two to three leaves stage of growth, the seedlings are transplanted into 250.93 cm2 square deep pots containers SVD-250 (T.O. Plastics, Clearwater, MN) containers were filled with

SS#8-F2-RSi potting medium, “SunGrow” (Sun Gro Horticulture, Agawam. MA). After senescence, gladioli corms are harvested and cooled at 2°C in darkness (Widmer 1958) for

>1,000 hours until they are ready to replant for further cycling.

Measurements. The number of weeks the gladioli seeds took to germinate from seed sowing is measured. Next, the number of weeks each seedling took to flower is measured from seed germination to anthesis. To measure reduced dormancy after senescence, the number of weeks is measured for plants to re-sprout seedlings while being in growing containers without a cold treatment.

Experiment design and statistical analysis. Seedling transplants were arranged in a completely randomized design (CRD) and all data were analyzed as unbalanced Analysis of Variance

(ANOVA). Tukey’s honestly significant difference (HSD) mean separations at P ≤ 0.05 using

JMP 14 statistical software (Campus Drive Cary, NC) were also performed.

2.4 Results

The gladiolus breeding for rapid generation cycling through selection of early flowering genotypes for 13 years has resulted in decreasing the long juvenile period (3 – 5 years) into second year flowering, and eventually reached a cycle 1 gladiolus in 2015. This resulted in the

RGC-1 with 20 plants which are cycle 1 plants of the total 3,903 plants across all grown populations, making the percentage of cycle 1 plants equal to 0.51%. The RGC-1 population is the first generation achieving rapid generation having plants able to flower from seed in 1st, 2nd

and latest 3rd year with a skewness of 1.48, in comparison to the starting original parental genotypes gladiolus ‘Carolina Primrose’ and ‘Atom’ which were able to flower from seed at the earliest in the 3rd year and the latest in the 5th year (Figure 4). The height and structure of the

RGC-1 gladioli are petite with a few florets (Figure 5).

In the recurrent selection generation (RGC-2), the number of cycle 1 plants are 8 out of

311 seedlings. However, this increased the cycle 1 plants by five-fold: 2.57% (Figure 6 and

Figure 7). The self-pollinated cycle 1 of RGC-2 are represented in the RGC-3 generation. These inbred lines resulted in 11 plants which are cycle 1 out of 318 plants with 3.46% of cycle 1

(Figure 8 and Figure 9). In a phenotypic comparison, the RGC-2 (recurrent selection hybrids) has more vegetative and reproductive growth than the inbred RGC-3 population (Figure 7 and Figure

9). A comparison of means in cycle for the RGC-1, RGC-2 and RGC-3 generations in an

ANOVA test (Table 1) shows a significant number of weeks that it took to germinate (p-value of

0.0156), while Tukey’s HSD means are significant. The RGC-2 generation mean is 4.5 weeks to germinate while the inbred generation (RGC-3) is intermediate with a mean of 3.91 weeks and the significantly earliest is RGC-1 with 2.65 weeks (Table 1). The number of weeks for all germinated cycle 1 plants ranged from 2 to 8 weeks, which falls within the range of gladiolus commercial seed germination of 3 weeks (Randhawa and Mukhopadhyay 1986). Line 16GL-2-

G7-1 in RGC-2 recurrent selection is the fastest flowering gladiolus among all populations with

12 weeks of seed germination, having noticeably short height, 3 leaves and barely stand erect

(Figure 10a). In the number of weeks germinated seedlings took to flower, the ANOVA test resulted in significant difference with p-value of <0.0001, while Tukey’s HSD means are significantly different as well The RGC-2 generation is the significantly earliest to reach flowering with a mean of 21.63 weeks, followed by the RGC-1 generation at 31.60 weeks, and the inbred generation being the latest with 39.00 weeks (Table 1). Therefore, the alternative 40

hypothesis (µ1 ≠ µ2 ≠ µ3) is accepted. The RGC-2 generation and RGC-3 had higher sprouting of senesced seedlings than the RGC-1 generation. In table 2, the RGC-2 generation has both recurrent selection hybrid full-sibs and inbred of RGC-1 (RGC-2 inbred). RGC-2 recurrent selection hybrid has 54.98% re-sprouted plants out of 311 germinated seedlings, while the RGC-2 inbreds have 20.18% re-sprouted plants out of 555 germinated seedlings; the Chi-square shows highly significant differences in re-sprouting of dormant gladioli.

2.5 Discussion

The effort of 20 years of breeding and selection greatly shifted the perennial nature of gladiolus from five years (seed to flower) into an annualized gladiolus (Wilkins and Anderson,

2006). Furthermore, the fastest ever gladiolus to flower from seed in the history of gladiolus breeding is faster than G. praecox which was around ≈16-17 weeks (Roemer 1907). The cycle 1 flowering plants has a range of 3 to 6 leaves during flowering, the flower colors vary from creamy white color, orange peach, pink, and solid red. While the number of florets of cycle 1 plants ranges from single florets to five florets / stalk across all cycle 1 plants, they are short in height as well. However, after replanting the corm in its later cycles the plant height and number of florets increase as it reached peak corm and meristem size which is unlike the first-time the

RGC-1 plants flowered, which have petite structure and a smaller number of florets (Figure 10 and Figure 11). In addition, gladiolus cormels are normally grown for flowering stocks to maintain genotypes vegetatively (Singh 2006). However, cormels of RGC genotypes could flower from a 2 cm in circumference cormel (Figure 12), while the smallest recorded flowering- sized cormel was 3 cm in circumference (Fortanier 1973). Another feature observed in breeding is in the number of anthers and stigmas. Since gladiolus is a monocot plant having 3 anthers and 3

stigmas (Figure 13a), in line 13GL-35 and some its progenies have four anthers and stigmas

(Figure 13b), 5 anthers (Figure 13c), to 6 stigmas (Figure 13d). Gladioli commonly possess horizontal orientation of the florets (Figure 14a), but in the RGC 2 lines they showed an upward orientation position (Figure 14b), similar to a few wild gladioli species in South Africa (Goldblatt and Manning 1998).

For most of the cycle 1 plants, obtained through hybridization in recurrent selection, the

RGC trait could be a heterotic effect. Hybrid vigor is reported in gladiolus for plant height, number of days to corm sprouting and seed set (Hemanth Kumar 2008; Mahato 2015; Pokhrel

2012). In the case of cycle 1, the occurrence of cycle 1 plants from an inbred line contradicts the possibility of heterosis, which could rule out the heterosis theory, while inbreeding depression is a possible explanation for having low rate of inbred cycle 1 plants. Since most commercial gladioli are polyploid, this could further increase inbreeding depression even though most are self- compatible (Barringer 2007). However, gladiolus progenies of self-pollinated hybrids showed an increase in height over the original parents. Therefore, gladiolus do not show inbreeding depression in certain traits (McLean, Clark, and Fischer 1927). In Hippeastrum, tetraploid seedlings flower in 2 to 3 years while diploid in 2 years and F1 diploids flower from seed in 18 months. Thus, polyploids could delay seedling to flowering time (Meerow 2000), which is an additional disadvantage in breeding polyploids.

In their natural habitats, senesced gladiolus corms are dormant in the fall and winter seasons until dormancy break in the spring, followed by flowering in early summer. In the typical life cycle, gladiolus goes into deep dormancy and is not competent to re-sprout unless there was a cooling period to break dormancy (Kumar 2007). Whether flowering occurs or not, both scenarios end up in deep dormancy. However, in all the RGC populations there are a portion of

plants with the ability to sprout back after the vegetative growth period (Table 2). These plants can sprout under greenhouse growing conditions without lifting the corm from the containers and without a cold treatment (Figure 15). Although the ratio of cycle 1 plants is low, the seedling senescence and failure to flower, and then sprouting and continue growing vegetatively is increasing in every generation.

The increase in re-sprouted plants was higher in the recurrent selection plants when compared with the inbreds, indicating a possible heterosis effect (Table 2). Also, since the re- sprouted plants may or may not lead to flowering can indicate that due to failure to increase cycle

1 there are still promising leads as these “cycle 1.5” are the intermediate of phenotype that could generate a cycle 1 with further breeding and selection. These cycle 1.5 plants flower in their first year (<52 weeks) from seed germination and, thus, a cycle 1 based on cultivation but botanically it’s a cycle 2 plant because, there a new daughter corm developed on top of the mother corm (the original mother corm is formed right after germination) (Figure 16). Thus, the major differences of a typical cycle 2 cultivated gladiolus and a cycle 1.5 is the short dormancy which leads to the ability to sprout and flower in the first year while a cycle 2 has deep dormancy, requiring cooling to break dormancy and flower in the following year. Dormancy is a typical pattern in the gladiolus life cycle. Furthermore, cycle 1.5 is more vigorous in vegetative growth and floral characteristics (floret size and number) which is similar to cycle 2 in growth and development of vegetative and floral attributes. This can support the claim that cycle 1.5 is a cycle 2 in phenotype but the flowering occurs in the first year. This feature of “reduced dormancy” has never been reported in gladiolus since most gladiolus species are deciduous and senesce after the growing season with the notable exception of G. sempervirens which is the only evergreen gladiolus species (Goldblatt and Manning 1998)

This new phenomena in RGC gladiolus could be a closer step to breed gladiolus to be more of a subtropical geophyte with minimal dormancy, such as Hippeastrum (Kamenetsky

2012). Hippeastrum is a subtropical geophyte genus native to South America with mild winters; their growth in the wild is almost evergreen, while in greenhouse they can stay evergreen (Dole and Wilkins, 1999). The breeding of gladiolus through intensive selection of early germination, early flowering and constant greenhouse conditions could have resulted in selection for reduced dormancy genotypes. To overcome dormancy in gladiolus, one possible hypothesis would be to include the Gladiolus sempervirens species in the breeding program. Gladiolus sempervirens is known as the evergreen gladiolus with red scarlet flowers (Goldblatt and Manning 1998); an evergreen gladiolus species would provide the essential genes to eliminate dormancy during the summer season as commercial gladiolus senesce in late summer. Ideally, a commercial gladiolus with the evergreen trait could be a useful feature in garden gladioli for subtropical climates, adding a new value to further use gladiolus in home gardening and expand the potential outcome, and use of the rapid cycling generation technique is a vital tool to reach this goal.

In conclusion, rapid generation cycling is a tool to speed up the breeding objective, especially for long juvenile geophytes such as Tulipa and Narcissus (Fortanier 1973). The seed propagated gladioli, which are able to flower without a vernalization requirement, are similar to seed-propagated Lilium formosanum and L. ×formolongi hybrids (Anderson 2009). Also, it can be a feature to produce the phenotype of annual plant form which can be marketable, even under a gibberellin inhibitor treatment, since the RGC population proved to flower while commercial genotypes fail to flower significantly (Aljaser and Anderson 2020). In addition, non-vernalized gladiolus cormels could flower under long days only (Aljaser, A. unpublished data), similar to a non-vernalized lily bulb (Lazare and Zaccai 2016). All the new and uncommon traits in RGC could possibly be hypothesized as linked traits. Genetic analysis for possible flowering genes for 44

cycle 1 are currently under investigation. Further selection and breeding lines for the desirable the ideotype in producing potted gladiolus such as breeding for germination in week 1, increasing number of florets up to seven in all progenies, increasing the floret size to decorative size (8.9 –

11.4 cm) (Okubo and Sochacki 2012) and reducing dormancy. An example of such an ideotype is in Figure 17. A United States Plant Utility Patent has been filed for protection of all these unique traits displayed in the cycle 1 flowering and nondormant corms, as derived from the RGC selection techniques (Anderson and Aljaser, 2019).

2.6 Literature Cited

Anderson NO, Frick J, Younis A, Currey C (2012) Heritability of Cold Tolerance (Winter

Hardiness) in Gladiolus xgrandiflorus. In: Abdurakhmonov IY (ed) Plant Breeding.

IntechOpen. London, pp 297-312

Anderson NO, Carter J, Hershman A, Houseright V (2015) Rapid generation cycling enhances

selection rate of gladiolus xhybridus. Acta Hort. 1087:429-435

Anderson NO (2019) Selection tools for reducing generation time of geophytic herbaceous

perennials. In XIII International Symposium on Flower Bulbs and Herbaceous Perennials

1237:53-66

Anderson NO (2019) Breeding for dwarf, winter-hardy Iris domestica, Blackberry lily

(Iridaceae). Acta Horticulturae, 1263, 275-281. (XXX IHC – Proc. Int. Symp. on Ornamental

Horticulture: Colour Your World) doi: 10.17660/ActaHortic.2019.1263.36

Anderson, N.O. and Aljaser, A. 2019. RAPID CYCLING GLADIOLUS. U.S. Utility Patent. U.S.

plant utility patent office, Document No. 14483-30011.00 - Specification – 4087289.

Anderton, E. W., & Park, R. (1989). Growing gladioli. Timber Press

Barringer B (2007) Polyploidy, inbreeding depression, and the evolution of mating systems in

flowering plants. Dissertation, Cornell University

Beal, A.C., Hottes, A. C. (1916) Gladiolus studies. New York state college of agriculture, New

York

Chauvin, JE, Hamann H, Cohat J, Le Nard M (1996) Selective agents and marker genes for use in

genetic transformation of Gladiolus grandiflorus and Tulipa gesneriana. Acta Hort. 430:291-

298

De Hertogh A, Le Nard M (1993) Physiology of flower bulbs. Elsevier, Amsterdam, Netherlands

Fortanier EJ (1973) Reviewing the length of the generation period and its shortening, particularly

in tulips. Scientia Horticulturae, 1(1):107-116

Griesbach RA (1972) The life-structure and function in gladiolus. The World of the Gladiolus.

North America Gladiolus Council. Edgewood Press, Edgewood, Maryland, pp 8-40

Goldblatt P, Manning J (1998) Gladiolus in Southern Africa. Fernwood Press, Vlaeberg, Cape

Town

Goldblatt P, Manning J, Dunlop G (2004) Crocosmia and Chasmanthe. Royal Horticultural

Society. Timber Press

Hemanth Kumar P, Kulkarni BS, Jagadeesha RC, Reddy BS, Shirol AM, Mulge R (2008)

Combining Ability and Heterosis for Growth Characters in Gladiolus (Gladiolus hybridus.

Hort). Karnataka Journal of Agricultural Sciences, 21(4):544-547

Kamenetsky R (2012) Biodiversity of Geophytes Phytogeography, Morphology, and Survival

Strategies. In: Kamenetsky R, Okubo H (eds) Ornamental geophytes: from basic science to

sustainable production. CRC press, Florida, pp 57-76

Kamenetsky R, Zaccai M, Flaishman MA (2012) Florogenesis. In: Kamenetsky R, Okubo H

(eds) Ornamental geophytes: from basic science to sustainable production. CRC press,

Florida, pp 197-232

Kumar P. N., Raju DVS (2007) Dormancy in gladiolus: The cause and remedy-A

review. Agricultural reviews-agricultural research communication Centre India, 28(4):309

Lazare, S., & Zaccai, M. (2016). Flowering pathway is regulated by bulb size in Lilium

longiflorum (Easter lily). Plant Biology, 18(4), 577-584

Mahato SK (2015). Genetic studies of some gladiolus genotypes and standar lization of invitro

seed germination. Doctoral dissertation, Uttar Banga Krishi Viswavidyalaya

McLean FT, Clark WE, Fischer EN (1927) The Gladiolus book. Doubleday Page & Co., New

York, p: 131

Meerow AW (2000). Breeding amaryllis. Callaway DJ, Callaway MB (eds) Breeding ornamental

plants. Timber Press, Portland, OR, 174-195.

North America Gladiolus Counsel (1999) Gladiolus parentages. North America Gladiolus

Counsel. http://www.gladworld.org/Glads1975-1999.pdf/. Accessed 23 May 2019

Okubo H, Sochacki D (2012) Botanical and horticultural aspects of major ornamental geophytes.

In: Kamenetsky R, Okubo H (eds) Ornamental geophytes: from basic science to sustainable

production. CRC press, Florida, pp 77-123

Poon TB, Pokhrel A, Shrestha S, Sharma SR, Sharma KR, Dev M B (2012) Influence of

Intervarietal and Interspecific Crosses on Seed Set of Gladiolus under Mid-hill Environments

of Dailekh Condition. Nepal Journal of Science and Technology, 13(1):17-24.

Randhawa GS, Mukhopadhyay A (1986) Floriculture in India. Allied Publishers, New Delhi

Rees AR (1992) Ornamental bulbs, corms and tubers. CAB international, Wallingford

Roemer F (1907) Gladiolus praecox. Neue Gladiolen-Klasse, deren Sämlinge im ersten Jahre

blühen. Möller’s deutsche Gärtn. Ztg. 6:64-66

Singh AK (2006) Gladiolus. In: Singh AK (ed) Flower crops: cultivation and management. New

India Publishing Agency, New Delhi, pp 147-166

U.S. Department of Agriculture. National Agricultural Statistics Service (2019) Floriculture 2018.

USDA Economics, Statistics and Market Information System, Albert R. Mann Library,

Cornell University, Ithaca, New York

Vaughn KV (2015) Breadless irises: a plant for every garden situation. Schiffer, Pennsylvania, pp

140

W. Atlee Burpee Company, Henry G. Gilbert Nursery and Seed Trade Catalog Collection (1913)

Burpee's annual 1914: the plain truth about seeds that grow. W. Atlee Burpee & Co.,

Philadelphia pp. 123

Widmer RE (1958) The Determination of Cold Resistance in the Garden Chrysanthemum and Its

Relation to Winter Survival. Proceedings of the American Society for Horticultural Science

71:537-546

Wilkins, H.F. and N.O. Anderson. (2006). Creation of new floral products. Annualization of

perennials—Horticultural and commercial significance. pp. 49-64. In: Anderson, N.O. (Ed.).

2006. Flower Breeding & Genetics: Issues, challenges, and opportunities for the 21st century.

Springer, Dordrecht.

Zlesak, D. and N.O. Anderson. (2009). Inheritance of non-obligate vernalization requirement for

flowering in Lilium formosanum Wallace. In: R. Kaminetsky (Ed.). Special Issue:

“Ornamental Geophytes”, Isr J of Plant Sci 57(4):315-327. 49

Table 1. Mean number of weeks of gladiolus seed germination, number of weeks of flowering from seed germination and percentage of flowering in every generation.

Generation n Weeks of germination Weeks of flowering Percentage of flowering

RGC-1 20 2.65b, z 31.60b 0.51c

RGC-2 8 4.50a 21.63a 2.57b

RGC-3 11 3.90ab 39.00c 3.46a p-value *, y *** ***

z Means within a column not followed by the same letter are significantly different at P≤0.05 using Tukey’s honestly significant difference (HSD) means comparison. y ANOVA for *, *** significant at P =0.05, 0.001, respectively.

Table 2. Number of gladiolus seedling, number of re-sprouted seedling, dormant and percentage of re-sprouted gladiolus plants in every generation and 1:1 Chi-square (χ2) test ratios of the tested Gladiolus generations.

Generation number of seedling Re-sprout Dormant Re-sprout%

RGC-2 recurrent Selection 311 171 140 54.98%

RGC-2 inbred 555 112 443 20.18%

D.f. 1

χ2 109.74 p-value ***, z

z, *** significant at P =0.001

A B

Figure. 1 A. University of Minnesota gladiolus breeding lines with unique flower colors and patterning. B. Yellow hybrid progenies having different red tips intensity, possible signs of an additive effect. Photos credits: Jaser Aljaser. 52

Figure 2. The breeding scheme for rapid generation cycling (RGC) gladiolus; the number of flowering plants in first year from seed to the total number of germinated plants.

Figure 3. The University of Minnesota gladiolus breeding lines have large capsule (fruit) as evidence of fertility, the seed/fruit set is 396/3.

3500 3196 3000 RGC-1 population 2500 plants 2000 1500

1000 Flowering 500 636 20 0 0 1 2 3 4 5 Cycles

Figure 4. Representation of number of flowering plants in cycles (years) in respect of parent generation (Green line) and RGC-1 population (Blue line). The parent generation represent the two parents used for breeding: G. ×hybridus ‘Atom’ to G. dalenii ‘Carolina Primrose’, the parents did not flower in first year of seed germination, the earliest year of flowering for parent generation is the cycle 3 (third year). While the blue line represent the result of selection and breeding of the parents for earlier flowering gladiolus, eventually reaching into the RGC-1 generation which was able to reach flowering in first year of seed germination, 20 plants able to flower in cycle 1 (first year), 3,196 plants flowered in cycle 2 (second year) while 636 plants flowered in cycle 3 (third year).

Figure 5. Phenotypes of the rapid generation cycling, cycle 1 (RGC-1) gladioli, gladiolus flowered in 24 weeks (left), gladiolus flowered in 22 weeks (right). Photo credit: David Hansen,

Minnesota Agricultural Experiment Station.

Flowering plants in RGC-2 recurrent selection 250 232

200

150

100 71 Flowering Flowering plants 50 8 0 0 1 2 3 Cycles

Figure 6. Frequency of the number of flowering plants in the RGC-2 recurrent selection generation. The total plants of RGC-2 are 311 germinated plants, the number of cycle 1 flowering gladiolus is 8 plants, cycle 2 flowering gladiolus 232 plants and cycle 3 flowering gladiolus 71 plants.

Figure 7. Phenotypes of the RGC-2 recurrent selection gladioli. Gladiolus flowered in 22 weeks

(upper left), 20 weeks (upper right) and 19 weeks (bottom). Photo credits: Jaser Aljaser. 58

Flowering plants in RGC-3 inbred population

350 307 300

250

200

150 RGC-3 inbred population

100 Flowering Flowering plants

50 11 0 1 2 Cycles

Figure 8. The number of gladiolus plants in rapid generation cycling (RGC-3) inbreds flowering in cycles 1 and 2.

Figure 9. Phenotypes of rapid generation cycling (RGC-3); all these gladiolus seedlings are inbreds. Gladiolus flowered in 44 weeks (upper left), 23 weeks (upper right) and 47 weeks

(bottom). Photo credits: Jaser Aljaser. 60

A B

Figure 10. A. Phenotypes of the fastest flowering gladiolus ever reported (seed to flower in 12 weeks) in Year 1. B. The identical genotype replanted in following season (Year 2), having more vegetative and reproductive growth. Photo credits: Jaser Aljaser.

Figure 11. Phenotypes of a rapid generation cycling (RGC-1) plant in its cycle 1 (left; Year 1) and the same genotype planted in second year (Year 2), having a higher number of florets due to being at peak meristem and corm size (right). Photo credits: David Hansen, Minnesota

Agricultural Experiment Station (left) and Jaser Aljaser (right).

Figure 12. Phenotype of a gladiolus seedling which flowered from a small cormel (2 cm in circumference) without a cooling treatment. Photo credit: Jaser Aljaser.

A B

C D

Figure 13. A. A classic wild type gladiolus having 3 anthers and 3 stigmas. B. Rapid generation cycling (RGC) hybrid gladiolus with 5 anthers; C. 4 anthers; D. 6 stigmas. Photo credits: Jaser

Aljaser.

A B

Figure 14. A. Cultivated gladiolus ‘Glamini’® florets are horizontal in position. B. A rare feature in rapid generation cycling (RGC) gladiolus with an upward floret position. Photo credits: Jaser

Aljaser.

Figure 15. Phenotype of a re-sprouting rapid generation cycling (RGC) gladiolus without a cooling treatment and after the end of its growing cycle (Year 1). Note that gladiolus grown in containers and reach the end of their cycle can be left in the container for a period of 4 weeks to dry then harvest the corm and cormels, at the end of the 4 weeks some RGC gladiolus re-sprouted and broke their typical deep dormancy. Photo credit: Jaser Aljaser.

Figure 16. The morphological differences between seed propagated rapid generation cycling

(RGC) cycle 1gladiolus (left) and a vegetatively propagated cormel which sprouted and then flowered (right). The RGC cycle 1 gladiolus has no daughter corm as it is germinated from seed and formed mother corm and flowered while the vegetatively propagated cormel has a mother corm (flattened) and a daughter corm above it has a flowering stalk. Photo credits: Jaser Aljaser.

Figure 17. Phenotype of a desired ideotype gladiolus for potted production. Photo credit: Jaser

Aljaser.

3.0 Effects of a gibberellin inhibitor on flowering, vegetative propagation and

production of rapid generation cycling gladiolus for potted plant production

To be submitted to HortScience

Jaser A. Aljaser1 and Neil O. Anderson2

Department of Horticultural Science, 1970 Folwell Avenue, University of Minnesota, St. Paul,

MN 55108 USA

1 Graduate Research Assistant 2 Professor; corresponding author email: [email protected] 69

Subject Category: Plant Growth Regulators

Effects of a gibberellin inhibitor on flowering, vegetative propagation and production of rapid generation cycling gladiolus for potted plant production

Additional index words. Gladiolus ×hybridus, corms, cormels, geophyte, gibberellin inhibitors, flowering, rapid generation cycling, cycle 1 flowering.

3.1 Preface

Gladiolus (Gladiolus ×hybridus Rodigas) is an asexually propagated, herbaceous perennial and an economically important cut flower crop. In commercial production, gladioli have tall flower stalks, which limit their use to cut flowers and annual garden plants. The gladiolus breeding program at the University of Minnesota has bred and selected rapid generation cycling (RGC) cycle 1 gladiolus, which can flower in <1 year from seed instead of the norm of 3-

5 years (which are vegetatively propagated as corms). Gibberellin inhibitors, such as ancymidol, are used as plant growth retardants to control height in potted plants and higher concentrations can inhibit flowering along with negative side effects. The aim of this study is to investigate the growth, flowering, and corm/cormel production response of cycle 1 gladiolus to the gibberellin inhibitor, ancymidol (0, 100 and 400 ppm soak) in comparison to non-cycle 1 genotypes and commercial cultivars for potted gladiolus production. Cycle 1 genotypes flowered with all ancymidol concentrations while non-cycle 1 genotypes had significantly less flowers or were completely non-flowering under higher levels. All tested genotypes had increased in the width of 70

leaves, as ancymidol concentration increased. Conversely, plant and flower stalk heights reduced as the ancymidol concentration increased while the number of stalks were non-significant. Corms, cormel number and fresh weights decreased in all genotypes except for one cycle 1 genotype, which had an increase in both corm number and fresh weight when treated with 100 ppm ancymidol. Cycle 1 gladiolus are more resilient to this gibberellin inhibitor even at high concentrations and can potentially be used for gladiolus potted plant production.

3.2 Introduction

Gladiolus (Gladiolus ×hybridus Rodigas) is member of the Iridaceae, native to South

Africa (Goldblatt and Manning, 1998), and a major cut flower in the floriculture industry (ranked in the top ten species). The 2015 wholesale farmgate value of cut flower gladiolus is $20M in the

U.S. (USDA-ARS, 2019) while in China, according to Forestry Bureau of China, the value is

~$41M in 2011 (Fukai, 2012). It is also commonly grown as an ornamental garden plant (non- hardy in northern latitudes, USDA Z3-4) (Anderson et al., 2012).

Due to its economic importance, the University of Minnesota flower breeding program is developing cold tolerant gladiolus for USDA Z3-4 (Anderson et al., 2012) with reduced generation cycling (Anderson et al., 2015). Breeding dwarf gladiolus for potted plant production and rapid generation cycling (RGC) seed-propagated, F1 hybrids are also breeding objectives and would be new traits for this crop. Rapid generation cycling with the ability to flower from seed in

<1 year (as early as 4-6 months) with a reduced juvenility and/or dormancy period is now a possibility due to 20 years of directed breeding and selection for earliness (Anderson et al., 2015); a U.S. Plant Utility Patent has been filed for breeding and selecting these phenotypes (Anderson

and Aljaser, 2019). Such seed-propagated hybrids could also be forced to flowering as vegetative clones without a cold treatment (Aljaser, 2020)

All gladiolus species are geophytes with corms (compressed stems) as underground storage organs (De Hertogh and Le Nard, 1993). In commercial production, gladioli are planted as mature (3-5 years old) corms (Dole and Wilkins, 2005). Gladioli are vegetatively propagated through daughter corms and cormels for commercial production. “Daughter corms” are defined as a specialized underground organ consisting of an enlarged stem axis with distinct nodes and internodes and enclosed by dry, scale-like leaves while cormels refer to small corms arising from a mother corm (De Hertogh, and Le Nard, 1993). Usually only one daughter corm is produced by the “mother corm” each year whereas the number of cormels produced per year varies among cultivars (Cohat, 1993). Asexual propagation by any means (division, scooping, scoring, tissue culture) that produces the highest number of propagules (cormels, corms) as fast as possible is critical for meeting market demands for a clonal gladiolus cultivar. The need to increase the number of daughter corms and cormels quickly increases propagule pressure. Tissue culture provides an alternative method to produce cormels (Simonsen and Hildebrandt, 1971), as well as the use of plant growth regulators such as gibberellins (GA), particularly GA3 and GA4+7, which are naturally occurring plant growth regulators involved in various roles such as seed germination, cell elongation and flower induction (Thomas and Hedden, 2006). Gibberellins can increase the number of gladiolus cormels by means of exogenous applications (Khan et al., 2011).

Flowering in gladiolus is governed by factors such as: corm size, vernalization, light intensity, long day photoperiod, ambient temperature and gibberellins (Ehrich, 2013;

Kamenetsky, et al., 2012). In geophytes such as Zantedeschia, storing the tubers resulted in a sharp increase in the endogenous gibberellin levels in the buds (Naor et al., 2008). Likewise, in

gladiolus corms, gibberellin levels increase during cold storage while abscisic acid (ABA), an inhibitor, decreases in concentration (Wu et al., 2015). Additionally, exogenous applications of gibberellins can hasten flowering (Tonecki, 1980; Sudhakar and Kumar, 2012).

Gibberellin inhibitors consist of wide range of plant growth retardants. Their primary use in commercial plant production is to reduce plant height through decreased cell elongation and cell division (Rademacher, 2016). Commercial applications are used to restrict plant height and establish a uniform height in ornamental crops using appropriate concentration of gibberellin inhibitors for bulbous crops, such as Hippeastrum (Miller et al., 2012).

Ancymidol is a pyrimidine analog plant growth regulator. Its mode of action is blocking the monoixygenase enzyme that catalyzes ent-kaurene oxidation, a necessary step in the pathway between ent-kaurene and gibberellins (Rademacher, 1991). Ancymidol or “A-Rest®” (SePRO

Corp., Carmel, IN) is applied at low rates ranging between 10 to 200 ppm for foliar sprays and

0.15 to 0.5 mg per 15.24 cm2 container for substrate drenches (Whipker and Evans, 2012). For gladiolus, a drench application rate of 1.5 mg A-Rest / 1.89 L is reported to effectively reduce height (Shaw et al., 1991). Conversely, for drench applications of other types of gibberellin inhibitor such as 0.8% 2-chloroethyltrimethylammonium chloride (CCC, chlormequat or

Cycocel; OPH, Mainland, PA) resulted in increases in stem length, the number of florets per spike and slightly later flowering dates (Halevy and Shilo, 1970). These plant growth regulator compounds have not been tested on newly developed winter-hardy and dwarf germplasm in the

University of Minnesota gladiolus breeding program, as well as new cycle-1 seed-propagated hybrids that flower in <1 yr. from sowing (Anderson et al., 2015).

The University of Minnesota Flower Breeding and Genetic gladiolus program has objectives of producing new phenotypes of gladiolus with rapid generation cycling whereby as

many as three cycles/year (where a cycle is equivalent to 1-year growth, dry down and vernalization treatments) enable faster breeding and selecting (Anderson et al., 2015; Anderson and Aljaser, 2019). In gladiolus, the period of seed to flower may encompass 3-5 years. Anderson et al. (2015) created a RGC program for gladiolus which included enhanced selection of seed- propagated hybrids for early germination, early leaf unfolding, high leaf numbers in RGC 1-3, early flowering stalk emergence, and hastened flowering (flower bud initiation and development).

All selection was done under standard growing conditions for the species, i.e. seed germination

(2-7 weeks) in glasshouse mist systems (21°C day/night) followed by subsequent growth of transplants for seven weeks in glasshouses (24/20°C day/night; inductive long day photoperiods for flowering, >150 μmol m-2 sec-1, 0600-2200 HR). Several genotypes were selected which flowered <1 year from seed in cycle 1, meaning they did not require a vernalization period to mobilize GA and breakdown ABA concentrations. Since these unique hybrids were bred and selected without vernalization treatments, it is unknown whether these genotypes have the same response to GA inhibitors based on published literature on geophytes. The objective of this study is to demonstrate cycle 1 gladiolus genotypic response to a GA inhibitor for flowering, vegetative propagation production (corm and cormels) and their overall use for potted plant production of gladiolus. The null hypothesis is: Ho = Treatment of ancymidol negatively influence gladiolus plant growth and flowering capability. The alternative hypothesis is: HA = Ancymidol concentrations positively impact in gladiolus plant growth and flowering capability.

3.3 Material and Methods

Plant material. Seven gladiolus genotypes (vegetative clones) were used in this experiment. Two genotypes were commercial cultivars, ‘Amsterdam’ and ‘Bananarama’, which

served as controls (comparisons). Flowering size corms of ‘Amsterdam’ and ‘Bananarama’ were obtained from Noweta Gardens (Table 1) during 2017 and had been produced in the same field conditions by the producer. ‘Amsterdam’ is a white colored gladiolus which is a hybrid cultivar derived from unknown parents bred by J. and P. Snoek and Sons, Ltd. (Flevoland, The

Netherlands) in 1992 (North America Gladiolus Counsel, 1999). ‘Bananarama’ is new yellow colored cultivar bred by Coöperatieve Kwekersvereniging “For Ever” U.A. (Sint Maarten, The

Netherlands) in 2013 (KAVB, 2014). Upon receipt of mature commercial corms with flowering capacity, they were cooled at 2°C in darkness (Widmer, 1958) for >1,000 hours prior to experimentation. Five University of Minnesota RGC gladiolus breeding lines were also tested:

RGC-Genotype 1, RGC-Genotype 2, RGC-Genotype 3, RGC-Genotype 4, RGC-Genotype 5

(Table 1). All RGC hybrid cormels were produced in the same greenhouse selection environment

(Anderson et al., 2015) and were capable of flowering. Thus, RGC cormels are physiologically equivalent to the mature commercial corms for flowering capacity. As many RGC hybrid cormels as possible were obtained and vernalized along with the commercial cultivars for the same duration prior to the commencement of this experiment.

Treatments. Three treatments (0, 100 and 400 ppm) of the gibberellin inhibitor “A-

Rest®” (SePRO Corp., Carmel. IN) or ancymidol: α-cyclopropyl-α-(p-methoxyphenyl)-5- pyrimidinemethanol (a.i. 0.0264%) were used (Table 1). All corms were soaked in the solutions for a period of 24 hours prior to planting. There were n=4-8 replications/genotype, depending on the availability of cormels from the RGC genotypes (Table 1). Thus, this experiment was an unbalanced design.

Greenhouse growing conditions. After 24 hours of treatment, corms were planted into

1679.776 cm2 square, deep pots (Belden Plastics, St. Paul. MN) in week 23 (2017) and grown for

18 weeks. Containers were filled with SS#8-F2-RSi potting soil, “SunGrow” (Sun Gro

Horticulture, Agawam. MA). The corms were grown in a long day photoperiod (0800 – 1600 HR supplied by 400-W high-pressure sodium lamps + 2200 to 0200 HR night interruption, >150

µmol m-2 sec-1) at a minimum setpoint of 18o C (day/night), 70-80% relative humidity, with irrigation accomplished using constant liquid feed (CLF) of 125 ppm N from water-soluble 20N–

4.4P–16.6K (Scotts, Marysville, OH) and deionized water on weekends. Standard fungicide drenches and insecticides were applied either monthly or as needed, respectively.

Measurements. Foliage height (cm) was recorded at the peak of highest growth

(measured from soil line to the tip of the uppermost leaf), flower stalk height (cm; measured as length from the uppermost leaf to the tip of the uppermost floral bud), flowering (+/-), and leaf width (cm) for widest leaf on each replicate. At the termination of the experiment (8 weeks after commencement), the number of corms and cormels were counted for production purposes and their respective fresh weights (g) recorded.

Statistical analysis. Replicates were arranged in a complete random design (CRD) and all quantitative data were analyzed as unbalanced Analysis of Variance (ANOVA) and Tukey’s honestly significant difference (HSD) mean separations at P ≤ 0.05 using JMP 13 statistical software (Campus Drive Cary, NC). Chi-squares (χ2) were calculated using a 1:1:1 χ2 test ratio for equal distribution of flowering response among the three different A-rest treatments (0, 100 and

400 ppm).

3.4 Results

Ancymidol treatments resulted in variation in flowering between genotypes (Table 2).

‘Amsterdam’ and ‘Bananarama’ did not flower even after 18 weeks from corm planting in pots in comparison to control, which flowered in 10 and 11 weeks, respectively (Table 3). Genotype

RGC-1 did not flower in all treatments, while RGC-2, RGC-3, RGC-4 and RGC-5 all flowered even at the high ancymidol concentration, yet with variation in number of flowering plants in each treatment (Table 2). All flowering genotypes were not significantly different in 1:1:1 Chi- square χ2 except for ‘Bananarama’ (Table 2). RGC-2, RGC-3, RGC-4 and RGC-5 genotypes, which flowered at high ancymidol concentration were delayed in flowering by 1-2 weeks from the control (Table 3).

Plant height ranged from 19 – 108 cm (Table 4). Ancymidol significantly decreased plant height in higher concentrations in all tested genotypes in exception of RGC-4 in which 100 ppm

(105.3 cm in height) was not significant to 0 ppm treatment (106.0 cm in height) (Table 4).

The number of stalks was not significantly influenced by the treatment, but genotypes were significantly different (Table 4). Leaf width was significantly increased with ancymidol treatment, ranging from 1.5 to 5.6 cm (Table 4).

Corms and cormels were harvested after all plants senesced (dry down of 18 weeks after termination of the experiment). They ranged in number from 1.0 to 8.0 with mean fresh weight of

3.7 to 40.8 g (Table 4).

3.5 Discussion

Gibberellin inhibitors such as ancymidol and paclobutrazol are reported to delay flowering in Tulipa (McDaniel, 1990), Lilium (Bailey and Miller, 1989) and Gladiolus (Ahmad et al., 2014). However, high concentrations of ancymidol resulted in completely inhibiting flowering in ‘Bananarama’ (Figure 1) as it was the only genotype that flowered in the 0 ppm

(control) treatment with complete lack of flowering in the other two treatments (100 and 400 ppm

A-rest). This was also reported in different Watsonia species, as dipping the corms in 0.5, 1 and 2 mg of paclobutrazol produced non-flowering plants while post-planting applications of 5, 10 and

25 of paclobutrazol resulted in flowering and marketable potted plants (Ascough et al., 2006).

While treatments of exogenous gibberellins after pretreatment with paclobutrazol in Tulipa reversed the influence of paclobutrazol and promoted flowering (Rebers, Romeijn and Knegt

1994). This demonstrates the importance of gibberellins in the geophytic flowering pathway

(Naor et al., 2008). Furthermore, RGC-2, RGC-3, RGC-4 and RGC-5 genotypes, which flowered at high ancymidol concentration were delayed in flowering by 1-2 weeks from the control (Table

3). This been demonstrated using paclobutrazol on gladiolus which resulted in delayed flowering as the concentration increased (Laubscher et al., 2008) However, the RGC-3, RGC-4 and RGC-5 genotypes are different since they are bred to be cycle 1 gladiolus with flowering in less than one year from sowing, all of which show short dormancy (Aljaser, 2020). These are genotypically difference in response to commercial gladioli with deep dormancy rest (Kumar and Raju, 2007) overcome only with a cooling period. Pooled data for all genotypes were significantly different in flowering response (did not fit the 1:1:1 Chi-Square) as well for pooled non-cycle 1 genotypes

(‘Amsterdam’, ‘Bananarama’, RGC-1 and RGC-2) (Table 2). However, no significant differences were found with ¾ cycle 1 genotypes (RGC-3, RGC-4 and RGC-5), indicating a complete lack of effect from the ancymidol treatment. This difference among RGC genotypes is most likely 78

genetic in nature, although further testing would be required to determine the exact causal gene(s). Nonetheless, the parents and RGC3 – RGC5 are novel genotypes in gladiolus and will be used to enhance development of dwarf, seed-propagated cycle 1 selections that flower in <1 year from sowing. The reduction of plant height (Table 4) matches the reports of gladiolus using flurprimidol (Ahmad et al., 2014). Since gibberellin inhibitors are used to control the plant height by reducing the gibberellins biosynthesis, this leads to limiting cell expansion and, thus, affecting plant height (Cosgrove and Sovonick-Dunford, 1989). The application of ancymidol, paclobutrazol and uniconazole decreased the lily plant height even with 1-minute dip at different concentrations (Miller et al., 2002). In addition to plant height, the flower stalk height is also decreased by ancymidol in ‘Amsterdam’, RGC-3, RGC-4 and RGC-5, while the flower stalk height of RGC-4 wasn’t influenced by ancymidol treatments and the height was almost identical in all treatments. RGC-3 and RGC-5 the 400 ppm ancymidol concentration resulted in more marketable gladiolus for potted production in term of plant height and flowering capability

(Figure 2). In freesia, Freesia × hybrida, dipping corms in 200 ppm of ancymidol was also reported to reduce flower stalk height (Berghoef and Zevenbergen, 1989). In genotype RGC-1 and RGC-2, the number of stalks decreased significantly as ancymidol concentration increased, while only RGC-4 showed an increase in the number of stalks. However, low concentrations (10 and 20 ppm) of paclobutrazol on Ornithogalum thyrsoides (a bulbous geophyte) increased the number of spikes per plants (Banswal, 2012). This indicates that increased levels of gibberellin inhibitors may influence the number of emerging stalks by genotype and species level.

Leaf width was significantly increased with ancymidol treatment, ranging from 1.5 to 5.6 cm (Table 4). Ancymidol at 100 ppm resulted in the leaves with widest length which were higher than both the control and 400 ppm treatments in all genotypes with exception of RGC-2 (Table

4). This increase in leaf width doesn’t necessarily means a corollary increase in leaf area, as 79

reported by Bailey and Miller (1989) gibberellin inhibitors tend to reduce whole leaf area with increased applied concentration in Lilium longiflorum ‘Nellie White’ as well Freesia × hybrida

(De Hertogh and Milks, 1989).

The number of corms showed variation with respect to treatment, 100 ppm treatment significantly increased the number of corms only in RGC-4 and RGC-5, while the weight of corms did not increase the as control resulted in the highest fresh weight of all genotypes (Table

4). Similar results were reported for gladiolus (Ahmad et al., 2014). On the other hand, the number of cormels and fresh weight did not increase in number and weight. Likewise, the number of cormels decreased in those genotypes producing cormels. The decrease in number of harvested cormels and fresh weight was similarly to previous reports of gibberellin inhibitors in gladiolus, where concentrations as low 10 mg/L paclobutrazol were reported to decrease cormel formation in expense to increase corm swelling in tissue culture (Steinitz and Lilien-Kipnis, 1989). This could mean the corm-cormel relationship is quantitative vs. qualitative, as gibberellin inhibitor influence a reduction in the number of cormels and corms. However, a recent study on corm and cormel formation was linked to the GhAGPL1 gene, indicating the role of ADP-glucose pyrophosphorylase (AGPase) in starch accumulation, since silencing GhAGPL1 resulted in a reduction of corms and cormels (Seng and Wu et al., 2017). Therefore, ancymidol’s role as a gibberellin inhibitor could be hypothesized to interrupt the starch accumulation in corms and cormels, thus resulting in reduction in weight. Future studies will be directed to answer this question.

In conclusion, gibberellin inhibitors such as ancymidol should be applied at precise concentrations for each genotype, as higher concentrations could result in failure to flower and reduce the fresh weight of both corms and cormels, which are essential for gladiolus floral

production in the market. The tested cycle 1 gladioli have an increased tolerance of higher concentrations of gibberellin inhibitors, such as ancymidol, to reduce plant height. Yet, such RGC gladioli are still able to flower, unlike non-cycle 1 gladiolus genotypes, which exhibit reduced height and significantly less flowering capability at higher gibberellin inhibitor concentration

(Figure 3). Therefore, the recommended ancymidol concentration for non-cycle 1 gladiolus should not exceed 100 ppm, whereas cycle 1 gladiolus could tolerate as high as 100 ppm for potted gladiolus production. Further research will be conducted at lower concentrations to determine the recommended concentration to achieve the targeted produce.

Future studies should include measuring the rachis distance between florets, number of florets, whole leaf area, and histological cross sectioning of floral differentiation at the three leaf stage to determine if ancymidol treatments inhibited floral differentiation growth in non- flowering genotypes as gladiolus is reported to have visible floral spike at the three leaves stage

(Streck et al., 2015). The applied concentrations are relatively high (100 and 400 ppm); thus, lower concentrations would be required to study the influence of A-Rest.

3.6 Literature Cited

Ahmad, I., B.E. Whipker, and J.M. Dole. (2014). Flurprimidol preplant corm soaks or drenches

affect potted gladiolus production. Acta Hort. 1104:103-108.

Aljaser, J.A. 2020. Gladiolus breeding for rapid generation cycling for potted production and the

discovery of gladiolus genes, UFC and FLX. University of Minnesota, Minneapolis, U.S.A.

Ph.D. Dissertation.

Anderson, N.O., J. Frick, A. Younis, and C. Currey. 2012. Heritability of cold tolerance (winter

hardiness) in Gladiolus xgrandiflorus, p. 297-312. In: I.Y. Abdurakhmonov (ed.). Plant

Breeding. InTech.

Anderson, N.O., J. Carter, A. Hershman, and V. Houseright. 2015. Rapid generation cycling

enhances selection rate of Gladiolus xhybridus. Acta Hort. 1087:429-435.

Anderson, N.O. and A. Aljaser. 2019. RAPID CYCLING GLADIOLUS. U.S. Utility Patent. U.S.

plant utility patent office, Document No. 14483-30011.00 - Specification – 4087289.

Ascough, G.D., N.P. Mtshali, D.I. Thompson, N.O. Anderson, J.E. Erwin, and J. Van Staden.

2006. Watsonia. Floriculture, ornamental and plant biotechnology 4:559-605.

Bailey, D.A. and W.B. Miller. 1989. Whole-plant response of Easter lilies to ancymidol and

uniconazole. J. Amer. Soc. Hort. Sci. 144(3):393-396.

Banswal, A.K. 2012. Effect of bulb sizes, growing substrates and paclobutrazol doses in potted

Chincherinchee (Ornithogalum thyrsoides Jacq.). Dr. Y. S. Parmar University of Horticulture

and Forestry, Solan, India. MS Thesis, Table 7.

Berghoef, J. and A.P. Zevenbergen. 1989. The effect of precooling, environmental factors and

growth-regulating substances on plant height of freesia as pot plant. Acta Hort. 266 :251-258.

Cohat, J. 1993. Gladiolus, p. 297-320. In: A. De Hertogh and M. Le Nard (eds.). The physiology

of flower bulbs: A comprehensive treatise on the physiology and utilization of ornamental

flowering bulbous and tuberous plants. Elsevier.

Cosgrove, D.J. and S.A. Sovonick-Dunford. 1989. Mechanism of gibberellin-dependent stem

elongation in peas. Plant Physiology 89(1):184-191.

De Hertogh, A. and M. Le Nard. 1993. The physiology of flower bulbs: A comprehensive treatise

on the physiology and utilization of ornamental flowering bulbous and tuberous plants.

Elsevier.

De Hertogh, A.A. and R. Milks. 1989. Forcing Dutch-grown Freesias as potted plants in the US

and Canada. Acta Hort. 266:115-122.

Ehrich, L. 2013. Flowering in South African Iridaceae, p. 248-269. In: Ramawat, K.G. and J.M.

Merillon (eds.). Bulbous Plants: Biotechnology. CRC Press.

Fukai, S. 2012. Geophyte Research and Production in East and Southeast Asia, p. 421-434. In:

Kamenetsky, R. and H. Okubo (eds.). Ornamental geophytes: from basic science to

sustainable production. CRC press.

Goldblatt, P. and J. Manning. 1998. Gladiolus in Southern Africa. Fernwood Press (Pty) Ltd..

Halevy, A.H. and R. Shilo. 1970. Promotion of growth and flowering and increase in content of

endogenous gibberellins in Gladiolus plants treated with the growth retardant

CCC. Physiologia Plantarum 23(4):820-827.

Kamenetsky, R., M. Zaccai, and M.A. Flaishman. 2012. Florogenesis, p. 197-232. In:

Kamenetsky, R. and H. Okubo (eds.). Ornamental geophytes: from basic science to

sustainable production. CRC press.

KAVB. 2014. Registraties 2013, Hillegom. 14 November 2018

Khan, F.N., M.M. Rahman, M.M. Hossain, and T. Hossain. 2011. Effect of benzyl adenine and

gibberellic acid on dormancy breaking and growth in gladiolus cormels. Thai J. of

Agricultural Science 44(3):165-174.

Kumar, P.N. and D.V.S. Raju. 2007. Dormancy in gladiolus: The cause and remedy-A

review. Agricultural reviews-agricultural research communication Centre India 28(4):309.

McDaniel, G.L. 1990. Postharvest height suppression of potted tulips with

paclobutrazol. HortScience 25(2):212-214.

Milandri, S.G., C.P. Laubscher, and P.A. Ndakidemi. 2008. Hydroponic culture of Gladiolus

tristis: Application of paclobutrazol for flowering and height control. African Journal of

Biotechnology, 7(3).

Miller, C.T., C.M. Filios, and W.B. Miller. 2012. Effects of flurprimidol, paclobutrazol and

uniconazole soaks and drenches on amaryllis (Hippeastrum) growth and development. Acta

Hort. 1002:431-438.

Naor, V., J. Kigel, Y. Ben-Tal, and M. Ziv. 2008. Variation in endogenous gibberellins, abscisic

acid, and carbohydrate content during the growth cycle of colored Zantedeschia spp., a

tuberous geophyte. J. of plant growth regulation 27(3):211.

North America Gladiolus Counsel. 1999. Gladiolus parentages, Shoshone. 14 November 2018

Rademacher, W. 1991. Inhibitors of gibberellin biosynthesis: applications in agriculture and

horticulture, p. 296-310. In: Takahashi, N., B.O. Phinney, and J. MacMillan (eds).

Gibberellins. Springer-Verlag, New York, NY.

Ranwala, A.P., G. Legnani, M. Reitmeier, B.B. Stewart, and W.B. Miller. 2002. Efficacy of plant

growth retardants as preplant bulb dips for height control in LA and oriental hybrid

lilies. HortTechnology, 12(3):426-431.

Rebers, M., G. Romeijn, E. Knegt, and L.H. Van der Plas. 1994. Effects of exogenous

gibberellins and paclobutrazol on floral stalk growth of tulip sprouts isolated from cooled and

non‐cooled tulip bulbs. Physiologia Plantarum 92(4):661-667.

Schwab, N.T., N.A. Streck, C.C. Becker, J.A. Langner, L.O. Uhlmann, and B.S.M.R. Ribeiro.

2015. A phenological scale for the development of Gladiolus. Annals of Applied

Biology, 166(3): 496-507.

Seng, S., J. Wu, J. Liang, F. Zhang, Q. Yang, J. He, and M. Yi. 2017. Silencing GhAGPL1

Reduces the Quality and Quantity of Corms and Cormels in Gladiolus. J. Amer. Soc. Hort.

Sci. 142(2):119-125.

Shaw, P.M., K.A. Schekel, and V.I. Lohr. 1991. Height Control in Pot-grown Wood Violet'

Gladiolus Using Ancymidol. HortScience, 26(8):1089-1089.

Steinitz, B. and H. Lilien-Kipnis. 1989. Control of precocious gladiolus corm and cormel

formation in tissue culture. J. of plant physiology, 135(4):495-500.

Sudhakar, M. and S.R. Kumar. 2012. Effect of growth regulators on growth, flowering and corm

production of gladiolus (Gladiolus grandiflorus L.) cv. White friendship. Ind. J. Plant

Sci, 1(2/3):133-136.

Thomas, S.G. and P. Hedden. 2006. Gibberellin metabolism and signal transduction, p. 147-184.

In: P. Hedden and S.G. Thomas (eds.). Annual Plant Reviews Volume 24: Plant Hormone

Signaling. Blackwell Publishing.

Tonecki, J. 1980. Effect of growth regulators on shoot apex differentiation and changes in sugars

and free amino acids in gladiolus (gladiolus × hortorum cv. acca laurentia). Acta Hort.

109:347-356.

U.S. Department of Agriculture. National Agricultural Statistics Service 2019. Floriculture 2018.

USDA Economics, Statistics and Market Information System, Albert R. Mann Library,

Cornell University, Ithaca, New York

Whipker, B.E. and M.R. Evans. 2012. Regulation of plant growth, p. 373-390. In: P.V.

Nelson (ed.). Greenhouse Operation and Management. 7th ed. Prentice Hall, Upper Saddle

River, NJ.

Widmer, R.E. 1958. The Determination of Cold Resistance in the Garden Chrysanthemum and Its

Relation to Winter Survival. Proceedings of the American Society for Horticultural Science

71:537-546.

Wu, J., S. Seng, J. Sui, E. Vonapartis, X. Luo, B. Gong, C. Liu, C. Wu, C. Liu, F. Zhang, J. He,

and M. Yi. 2015. Gladiolus hybridus ABSCISIC ACID INSENSITIVE 5 (GhABI5) is an

important transcription factor in ABA signaling that can enhance Gladiolus corm dormancy

and Arabidopsis seed dormancy. Frontiers in plant science, 6:960.

Table 1. Number of replicates / treatment (0, 100, 400 ppm A-Rest®) and commercial or breeding source of the tested Gladiolus genotypes.

No. of replicates / treatment

of A-Rest® concentration (ppm)

Genotype 0 (Control) 100 400 Source

‘Amsterdam’ 8 8 8 Noweta Gardens, Inc. Three Rivers, MI

‘Bananarama’ 8 8 8 Noweta Gardens, Inc. Three Rivers, MI

RGC-1 4 4 4 University of Minnesota

RGC-2 4 4 4 University of Minnesota

RGC-3 7 7 7 University of Minnesota

RGC-4 4 4 4 University of Minnesota

RGC-5 5 5 5 University of Minnesota

No. of flowering plants in A-Rest® (ppm)

Genotype n 0 (control) 100 400 1:1:1 χ2 Significance

‘Amsterdam’ 24 4 1 0 5.20 NSz

‘Bananarama’ 24 5 0 0 9.98 **

RGC-1 12 0 0 0 ---y ---y

RGC-2 12 2 2 1 0.40 NS

RGC-3 21 6 1 5 3.50 NS

RGC-4 12 4 4 2 0.80 NS

RGC-5 15 5 1 3 2.67 NS

Pooled 120 26 9 11 11.26 **

Non-cycle 1 72 11 3 1 11.20 **

Cycle 1 48 15 6 10 3.94 NS

z NS, ** Non-significant and significant at P =0.01, respectively. y Not estimable due to no flowering in all replicates. 88

Genotype n treatments No. of flowering plants Mean no. of weeks reached to flowering

‘Amsterdam’ 8 0 4 10

8 100 1 12

8 400 0 ---z

‘Bananarama’ 8 0 5 11

8 100 0 ---

8 400 0 ---

RGC-1 4 0 0 ---

4 100 0 ---

4 400 0 ---

RGC-2 4 0 2 11

4 100 2 12

4 400 1 13

RGC-3 7 0 6 10

7 100 1 11 89

7 400 5 12

RGC-4 4 0 4 11

4 100 4 12

4 400 2 13

RGC-5 5 0 5 10

5 100 1 11

5 400 3 12

Treatment ---y

Genotype ***x

Treatment x genotype ***

z Not estimable due to no flowering in all replicates y Not estimable due to the number of non-flowering genotypes x ANOVA for NS, *** Non-significant and significant at P =0.001, respectively.

Table 4. Mean plant height (cm), flower stalk height (cm), number of stalks, leaf width (cm), no. of corms, fresh weight (FW) of corms (g), number of cormels, and fresh weight (FW) of cormels (g) for seven gladiolus genotypes corms treated with different concentrations of A-Rest®

(0, 100, and 400 ppm).

A-Rest® Plant Flower stalk No. of Leaf No. of FW of No. of FW of Genotype treatments height (cm) height (cm) stalks width (cm) corms corms (g) cormels cormels (g) ‘Amsterdam’ 0 102.8 abz 101.3 abc 1.7 bc 4.4 bcde 1.8 c 22.9 bcde 0.0 b ---x

100 78.0 cde 80.0 abc 2.0 bc 5.6 a 1.9 c 18.5 de 0.0 b ---x

400 52.4 fg ---y 1.6 bc 4.9 abcd 2.0 c 8.4 e 0.0 b ---x

‘Bananarama’ 0 106.3 a 107.8 ab 2.1 bc 3.5 efgh 2.1 c 28.1 abcd 0.0 b ---x

100 62.9 ef ---y 2.0 bc 5.2 ab 2.0 c 19.3 cde 0.0 b ---x

400 40.7 g ---y 1.7 bc 5.1 abc 1.8 c 17.1 de 0.0 b ---x

RGC-1 0 101.5 abc ---y 3.0 abc 2.2 hi 3.0 bc 29.2 abcd 2.5 b 1.4

100 53.3 efg ---y 2.0 bc 2.3 hi 2.3 c 14.2 de 0.8 b 0.2

400 19.0 g ---y 1.0 c 1.5 hi 1.0 c 3.7 e 0.0 b ---x

RGC-2 0 83.3 abcde 56.0 c 4.0 abc 2.6 fghi 3.3 bc 32.6 abcd 2.7 b 0.3

100 56.3 efg 55.5 c 2.8 bc 2.6 ghi 3.8 bc 17.2 de 0.0 b ---x

400 33.5 g 56.0 c 1.3 bc 2.2 hi 2.3 c 8.5 e 0.0 b ---x

RGC-3 0 108.0 a 121.2 a 1.3 bc 3.7 defg 1.3 c 26.1 abcd 24.0 a 4.1

100 75.7 cdef 107.0 abc 1.8 bc 4.4 bcde 2.0 c 26.1 abcd 7.5 b 1.6

400 67.0 def 65.0 c 1.0 c 4.1 cdef 1.5 c 21.2 bcde 2.8 b 0.9

RGC-4 0 106.0 a 113.0 ab 4.0 ab 3.7 defgh 3.5 bc 37.6 abc 1.3 b 0.2

100 105.3 a 98.5 abc 6.5 a 3.8 defg 8.0 a 40.8 a 0.8 b 0.1

400 89.0 abcd 82.5 abc 6.3 a 4.1 bcdef 6.5 ab 37.9 ab 0.0 b ---x

RGC-5 0 98.8 abc 85.6 abc 3.2 abc 3.0 fghi 3.3 bc 20.9 bcde 0.6 b 1.2

100 77.0 cdef 72.0 abc 3.3 abc 3.3 efghi 4.3 abc 8.9 e 0.0 b ---x

400 66.2 def 61.3 c 2.4 bc 3.1 fghi 2.0 bc 7.7 e 0.2 b 0.1

Treatment ***u ---w NS *** * *** ** ---v

Genotype *** *** *** *** *** *** *** ---v

Treatment x genotype *** NS NS *** NS NS *** ---v

z Means within a column not followed by the same letter are significantly different at P≤0.05 using Tukey’s honestly significant difference (HSD) means comparison. y Nonflowering x No production of cormels w Not estimable due to the number of non-flowering genotypes

Figure 1. ‘Bananarama’ gladiolus treated with A-Rest® concentrations 400, 100, 0 ppm, left to right respectively. Photo credit: Jaser Aljaser.

Figure 2. Rapid Generation Cycling (RGC) RGC-3 gladiolus treated with A-Rest® concentrations of 400, 100, 0 ppm, left to right respectively. Photo credit: Jaser Aljaser.

Figure 3. Gladiolus treated with A-Rest® concentration of 400 ppm. RGC-2 reached to flowering

(left), while ‘Bananarama’ remained in vegetative state (right). Photo credit: Jaser Aljaser

4.0 Discovery of UPSTREAM OF FLOWERING LOCUS C (UFC) and FLOWERING

LOCUS C EXPRESSOR (FLX) Genes in Gladiolus ×hybridus and G. dalenii

To be submitted to the Journal of Experimental Botany

Jaser A. Aljaser1, Neil O. Anderson2* and Andrzej Noyszewski3

Department of Horticultural Science, 1970 Folwell Avenue, University of Minnesota, St. Paul,

MN 55108 USA

1 Graduate Research Assistant

2 Professor; corresponding author email: [email protected]

3 Postdoctoral Associate

Subject Category: Breeding, cultivars, rootstocks, and germplasm resources

Discovery of UPSTREAM OF FLOWERING LOCUS C (UFC) and FLOWERING LOCUS C

EXPRESSOR (FLX) Genes in Gladiolus ×hybridus and G. dalenii

Additional index words. Gladiolus ×hybridus, flowering, geophyte, UFC, FLX, FLC, FRI

4.1 Preface

Gladiolus is a geophytic floricultural crop, cultivated for cut flower and garden ornamental uses. Ornamental geophytes such as gladiolus, lily, tulip and daffodil are examples of floral crops that are currently being investigated to understand the flowering pathway. While the environmental and hormonal factors leading to flowering are established in Arabidopsis.

However, the lack of genetic regulation is poorly understood. Thus, the importance of such an ornamental crop that relies on flowers (flowering) for economic purposes encourages researchers to discover the flowering genes to breed vigorous flowering cultivars. The understanding of the flowering mechanisms in the flowering pathway is also paramount. Herein we show the discovery of UPSTREAM OF FLOWERING LOCUS C (UFC) and FLOWERING LOCUS C EXPRESSOR

(FLX) genes in Gladiolus ×hybridus and G. dalenii. The UFC gene is adjacent to FLOWERING

LOCUS C (FLC) which is a floral repressor in many temperate species. FLX gene upregulates

FRIGIDA (FRI) which upregulates FLC expression. The discovery of both genes is a step forward in finding the FLC gene in gladiolus, provided they are linked. Seventeen gladiolus genotypes, consisting of early flowering and commercial cultivars, have the UFC gene, consisting 97

of four exons in two allelic forms. The UFC gene sequenced when translated into amino acid sequence and set in pair-alignment to other species, has up to 57% in amino acid identity to Musa acuminata. The UFC protein ranges in identity with pair-alignment to other species, reaching up to 57% in amino acid identity to Musa acuminata. The FLX gene in gladiolus has 3/5 (60%) exons in relative to Ananas comosus, i.e. lacking 2 exons and a partially complete gene sequence; the pair-alignment of the three exons shows up over all ~65% identity of FLX to Ananas comosus.

The UFC protein consists of a conserved domain, DUF966, which is higher in identity and pair-alignment, with up to 86% identity in Elaeis guineensis. The discovered FLX gene in gladiolus has 3/5 (60%) exons, i.e. lacking 2 exons and a partially complete gene sequence; the pair-alignment of the 3 exons shows up to ~65% of identity of FLX to Ananas comosus. These discovered two genes in gladiolus provide insight to further our understanding of the flowering and vernalization response in ornamental geophytes.

4.1 Introduction

“Florogenesis” is a process of flowering transitioning plant floral meristem from vegetative tissue to reproductive organs in angiosperms (Kamenetsky, Zaccai, & Flaishman,

2012). This transition is governed by flowering genes in which expression is influenced by factors such as vernalization, photoperiod, gibberellins, autonomous pathway and ambient temperature (Srikanth and Schmid 2011). In Arabidopsis, several flowering genes were discovered that are involved in flowering and act as floral integrators; some of these flowering genes are FT, SOC1, CO, VRN1, PPD, FCA, FLD, and FLK (Simpson & Dean 2002; Simpson,

2004). These floral integrator genes specifically upregulate flowering by promoting transition 98

from vegetative to flowering or repressing floral repressor genes. Repressor genes act as repressors of floral integrator genes and upregulates the expression of repressors, including genes such as FLC, FRI, FLX, VRN2, and SVP (Simpson & Dean 2002; Dean et al 2002; Ahn et al

2007).

FLC and FLC-like is floral repressors found in many dicotyledon plants, such as Malus

(Singh et al. 2016), Rosa (Zhang et al. 2017), Coffea (de Oliveira et al. 2014) and Brassica

(Peacock et al. 2001). The FLC gene is regulated by temperature changes throughout the year, both in annuals and perennials. In summer, FLC expression is upregulated through FRIGIDA

(FRI) by binding the FLC promoter through the DNA-binding protein SUPPRESSOR OF

FRIGIDA4 SUF4 (Choi et al. 2011). Also, FRI expression is upregulated by the EXPRESSOR OF

FLOWERING LOCUS C (FLX); both SUF4 and FLX are in FRI-specific pathway (Michaels et al.

2013). In winter, FLC is down-regulated through a process of vernalization as prolonged exposure of low temperature in winter in the meristem gradually reduces the expression of FLC

(Michaels and Amasino 1999). In addition to the vernalization pathway, the autonomous pathway reduces the expression of FLC both in the meristem and leaves (Michaels and Amasino 1999).

Gradual reduction of FLC allows FLOWERING LOCUS T (FT) to be expressed in the leaves and transported through phloem to reach to the meristematic tissue to stimulate the MADS box genes, which thereby induces flowering in Arabidopsis (Amasino 2005).

In wheat and barley, the flowering pathway is regulated by photoperiod, vernalization and the circadian clock (Turner 2013). VRN2 acts a flowering inhibitor and only through vernalization is the expression of it downregulated. Then, the floral integrator leads to flowering in winter wheat while spring wheat doesn’t require vernalization as the VRN2 gene is nonfunctional, although vernalization will speed up flowering in spring wheat, so vernalization

acts as a facultative stimulus (Dubcovsky 2004). In contract, maize and rice rely on plant age to build up sufficient energy requirements in order to transition to flowering through epigenetic of miR172 (Helliwell 2011). In monocotyledon geophytes (defined as herbaceous perennial plants with underground storage organs, e.g. bulbs, corms, tubers, etc., that promote winter survival) such as Gladiolus, Lilium, Tulipa, Narcissus and Crocus the flowering process is poorly understood. Factors of plant growth influencing flowering in commercial geophytes are well known (Ehrich, 2013) and include photoperiod, light intensity, autonomous pathway, gibberellins, ambient temperatures and cool temperatures (vernalization; Kamenetsky, Zaccai and

Flaishman, 2012). On the other hand, the clear genetic pathway is still in the early stages of the discovery and characterization, in comparison to the Arabidopsis model which can be applied to temperate dicotyledon plants (Fadón et al 2015). Only a few flowering genes have been discovered in geophytes (Kamenetsky, Zaccai and Flaishman, 2012), such as FT-like in Allium cepa (Taylor 2009; Taylor et al 2010), FT in Narcissus (Noy-Porat 2009), NLF in Narcissus

(Noy-Porat 2010) and LFY in Allium sativum (Rotem et al 2007; Rotem et al 2011), LFY in

Lilium (Wang et al 2008). Recently, many flowering genes have been discovered in Lilium

×formolongi, including FT, CO-like, AP2, GA1 and SOC1 (Jia et al 2017). The discovery of flowering genes in geophytes serve as valuable resources to draw the model pathway of flowering geophytes.

Geophytes such as Gladiolus, Lilium, Tulipa, Narcissus and Crocus are floricultural crops with ornamental value wherein flowering is essential to maintain the marketing value for these crops. Gladiolus ×hybridus Rodigas, commonly known as gladiolus, is commercially cultivated as a cut flower and as garden or landscape planta. Gladioli are geophytic plants with underground modified stems called corms, producing cormels as a means of vegetative

100

propagation (Cohat, 1993). Flower formation is a crucial step for its success as a cut flower.

Therefore, understanding the flowering pathway is vital to improve the floral market value.

Gladiolus has a genome size of 1100 Mbp, although it is unclear it is for haploid or diploid and the species is unknown (Kamo, Krens, & Ziv 2012), although the genome weight for gladiolus is recently measured, in G. communis 0.67-0.68 pg for monoploid G.s. (1Cx, pg) and G. italicus 0.61 pg for monoploid G.s. (1Cx, pg) (Castro, 2019). The limited knowledge in gladiolus genome is also reflected in lack of knowing gladiolus flowering genes, there is no flowering genes discovered in gladiolus except of gibberellin receptor gene GID1a in gladiolus (Luo and Lu et al 2016) yet the relationship of this gene with flowering isn’t established. In Arabidopsis, gibberellin binds to the gibberellin receptor forming GID1 complex that binds to DELLA and causing it’s degradation, enabling SOC1 and LFY to upregulate, leading to flowering in the gibberellin pathway (Blázquez & Weigel 2000; Lee et al 2003).

2006), barley (Monteagudo, 2019) and Brachypodium distachyon (Kaufmann, 2013). Although the FLC homologue in cereals did not discover any FRI gene, which upregulated FLC expression in Arabidopsis thaliana (Choi et al. 2011). A hypothesis to test would be that some monocots do not possess the flowering repressor FLC gene and rely on alternative gene(s) to acts as a repressor(s) miR172 through epigenetic in plant age-dependent of Zea mays and Oryza (Helliwell

2011). Yet the question remains whether there is FLC-dependent pathway in all monocot plants or whether monocots are independent of FLC.

101

The FLC gene is located between two flanking genes, UPSTREAM OF FLOWERING

LOCUS C (UFC; a gene found 4.7 Kb of upstream of FLC) and DOWNSTREAM OF

FLOWERING LOCUS C (DFC; found 6.9 Kb of downstream of FLC) in Arabidopsis (Finnegan et al 2004). UFC gene expression is repressed by vernalization, independent of FLC repression by vernalization (Finnegan et al 2004). Thus, both FLC and UFC are repressed by vernalization, yet both are not dependent on each other expression; the suppression is through chromatin modification in epigenetic manner (Finnegan et al 2004). The VRN1 gene is expressed with vernalization and acts as a floral integrator whereas the UFC gene is repressed and required by

VRN1 expression dependently (Sheldon et al 2009). The role for UFC in flowering is yet to be uncovered and it may not involve flowering to begin with, because vernalization only represses

UFC in seed while DFC is repressed by vernalization of the plant (Sheldon et al 2009). Insertion of the NPTII gene between the UFC and FLC region confirmed NPTII response to cold as the whole cluster region of FLC response to cold (Finnegan et al 2004). In UFC protein, a conserved domain DUF966 is present in Arabidopsis thaliana which has 92 amino acid its function is still unknown (Yoshida and Weijers et al 2019). This lack of knowledge in DUF966 function creates a challenge to identify the function of UFC protein. However, a recent study shows the role of UFC in Arabidopsis thaliana, the gene is designated as SOK2, and it would appear that it has a role in embryogenesis, root initiation, growth and branching of the primary and lateral roots (Yoshida and Weijers et al 2019). The conserved domain DUF966 is reported to be present in different species such as Oryza sativa, the OsDSR gene family contains DUF966 (Chengke & Lei 2017).

The ZmAuxRP1 gene which promotes the biosynthesis of indole-3-acetic acid (IAA) to increase resistance against pathogens in Zea mays (Ye et al 2019). The promoter for genes that contain

DUF966 have a defense-stress response to pathogens, Salicylic acid, Jasmonate acid, or drought or salinity (Ye et al 2019). Thus, all these genes that contain DUF966 vary in function but all of

102

them are triggered by environmental or stress stimuli to overcome an undesirable change influenced plant growth and development.

The FLX is a gene encoding a putative leucine zipper domain that are required for FRI- mediated activation of FLC in Arabidopsis (Dennis et al 2008) (Figure 1). Expression up- regulating FLC occurs in winter annual Arabidopsis (Lee and Amasino, 2013), while late flowering phenotypes exhibit strong expression of FLX. flx have early flowering which indicate a role of FLX in suppression of flowering (Dennis et al 2008). Several genes have been discovered in the FLX gene family, such as FLX-LIKE1 (FLL1), FLX-LIKE2 (FLL2), FLX-LIKE3 (FLL3),

FLX-LIKE4 (FLL4) (Choi et al 2011; Lee et al 2013). FLX and FLL4 are the most crucial genes in flowering time control in Arabidopsis (Lee and Amasino, 2013).

In order to test whether FLC is present in gladiolus, the adjacent gene (UFC) will also be searched for, along with FLX which is part of the FLC-dependent mechanism. Therefore, the objective of this study is to identify whether UFC and FLX genes occur in gladiolus germplasm present in the University of Minnesota Gladiolus Breeding Program. The null hypotheses tested are: Ho1 = There is no difference among gladiolus genotypes in the existence of the UFC gene; Ho2 = There is no difference among gladiolus genotypes in the presence of FLX.

4.2 Materials and Methods

Germplasm Tested. The number of gladiolus genotypes used for this study is 17 (Table 1) were chosen to represent a range of diversity within cultivated gladioli (both Gladiolus

×hybridus, and a wild species hybrid of G. dalenii) which includes nine genotypes of Rapid

103

Generation Cycling-1 (RGC-1; ones that flower in <1 year from seed; Anderson, 2015; Anderson,

2019; Aljaser and Anderson, 2020; Anderson and Aljaser, 2020) and eight genotypes of Non-

RGC (that require >1 to 5 years to flower from seed). Fourteen of these genotypes are parents and hybrids created by the University of Minnesota Gladiolus Breeding Program, while three additional genotypes are commercial cultivars. One genotype ‘Carolina Primrose’ is derived from the species G. dalenii (Table 1). All gladiolus pedigrees used in this experiment are published

(Aljaser and Anderson, 2020; Anderson and Aljaser, 2020) and commercial cultivars ‘Beatrice’

(an open-pollinated seedlings of unknown origin, occurring in a private garden, Brookfield,

Vermont, in 2003. This genotype was selected for its winter hardiness, surviving in USDA Z3).

‘Glamini’® a series of shorter in height than tall summer gladiolus, bred by Dutch breeders, bloom early, has a range of flowering colors (Wayside Gardens, 2020). ‘Carolina Primrose’ is an heirloom gladiolus, bred in 1908, yellow color flowers, collected at an old homesite in North

Carolina and it is a cultivar bred from G. primulinus (Oldhousegardens, 2020).

Greenhouse Environment. Mature (capable of flowering) gladiolus corms were planted into 1679.776 cm2 square, deep pots (Belden Plastics, St. Paul. MN) in week 23 (2017) and grown for 18 weeks. Containers were filled with SS#8-F2-RSi potting soil, “SunGrow” (Sun Gro

Horticulture, Agawam. MA). The corms were grown in a long day photoperiod (0800 – 1600 HR supplied by 400-W high-pressure sodium lamps + 2200 to 0200 HR night interruption, >150

µmol m-2 sec-1) at a minimum setpoint of 18o C (day/night), 70-80% relative humidity, with irrigation accomplished using constant liquid feed (CLF) of 125 ppm N from water-soluble 20N–

4.4P–16.6K (Scotts, Marysville, OH) and deionized water on weekends. Standard fungicide drenches and insecticides were applied either monthly or as needed, respectively.

104

DNA extraction and probe design. Newly expanded gladiolus leaves were harvested, placed in an ice box and sent to RAPiD Genomics® LLC (Gainesville, FL; http://rapid- genomics.com/home/) for DNA extraction, probe design, sequencing and computable analysis.

Probe designs for the UFC gene were based on banana, Musa acuminata subsp. malaccensis accession XM_009383889, from the GenBank Nucleotide Core (NCBI, 2016a) and oil palm,

Elaeis guineensis accession XM_010920607.2 (NCBI, 2017a). Probe design for FLX gene were based on oil palm, Elaeis guineensis accession XM_010924316.2 (NCBI, 2017b) and date palm,

Phoenix dactylifera accession XM_008801571.2 (NCBI, 2016b). The designed probe for UFC able to capture the locus in Musa acuminata and Elaeis guineensis by capturing the 2x coverage of the UFC exons in Musa acuminata and Elaeis guineensis. While the FLX probe capture the locus in Elaeis guineensis and Phoenix dactylifera. Then the probes able to amplify short read of

UFC and FLX genes in gladiolus. The reads are sequenced through Illumina dye sequencing technique, the raw data is demultiplexed using Illuminas BCLtofastq then assembled using

MaSuRCA® software (Zimin, et al., 2013) creating full assembly sequences scaffolds.

Afterword, read mapping using reference genome and blast to filter all assembled sequences for hits to the sequences provided for probes design (UFC and FLX), then count read numbers for each assembled sequence that passed the filter and accruing the final sequences for genetic analysis. Gene sequences will be deposited into GenBank.

Genetic Analysis. The sequence data for the UFC and FLX genes used in this study were found in the genetic sequence database under the following accession/ID numbers: Ananas comosus (Aco009327) UFC gene is from the Pineapple Genomics Database (Yu, Shi, Tang, Yang et al 2018); Musa acuminata (GSMUA_Achr5T28540_001) UFC from the Banana Genome Hub

(Droc et al 2013); Elaeis guineensis (p5.00_sc00099_p0095) UFC from the Malaysian Oil Palm

Genome Programme (Halim et al 2018); Asparagus officinalis 105

(evm.model.AsparagusV1_08.3493) UFC from the Asparagus Genome Project (Harkess et al

2017); Arabidopsis thaliana (At5g10150) UFC from The Arabidopsis Information Resource

(TAIR) (TAIR, 2020); Glycine max (Glyma.11G193000.1) UFC from the SoyBase (Grant et al

2010). The FLX protein was from the GenBank Nucleotide Core with accession numbers as follows: Ananas comosus (XP_020095672.1) (NCBI, 2017c), Musa acuminata

(XP_009420070.1) (NCBI 2016c), Elaeis guineensis (XP_010922618.1) (NCBI, 2019a),

Arabidopsis thaliana – FLX (NP_001154541.1) (NCBI 2019b), Arabidopsis thaliana – FLL1

(NP_566492.1) (NCBI 2019c), Arabidopsis thaliana – FLL2 (NP_001320766.1) (NCBI 2019d),

Arabidopsis thaliana – FLL3 (NP_564678.1) (NCBI 2019e), Arabidopsis thaliana – FLL4

(NP_001119474.1) (NCBI 2019f) and Glycine max (Glyma.15g269300) FLX was from the

SoyBase (Grant et al 2010).

Generated sequences were analyzed for gene prediction using the HMM-based gene structure prediction of FGENESH with Arabidopsis thaliana (Generic) as the specific gene- finding parameter. The predicted genes for UFC and FLX were analyzed in multi-alignment using

Geneious© software (Biomatters Ltd, Auckland, NZ). The UFC protein sequences of gladiolus were analyzed for conserved domains using the Protein Homology/analogy Recognition Engine

V 2.0 (Phyre2) browser (Kelley et al 2015). Then the alignment of conserved domain was formed to compare the matching and differences in each amino acid in the sequences of gladiolus and the other comparison species. A phylogenetic tree of all UFC genotypes of Gladiolus was formed by computing the distances using the Tamura-Nei method and were in the units of the number of base substitutions per site. The tree building used the Neighbor-Joining method and a bootstrap test was performed for each tree (500 replicates).

106

4.3 Results

The investigated genotypes of Gladiolus resulted from the gladiolus breeding program for the selection of rapid generation cycling, which includes 14 breeding genotypes and 3 commercial cultivars (‘Vista’, ‘Glamini®’ and ‘Carolina Primrose’) listed in Table 1. Rapid

Generation Cycling (RGC) in gladiolus is the ability of flowering in the first year or less form seed sowing (Aljaser and Anderson, 2020; Anderson and Aljaser, 2020). Typically, gladiolus are juvenile in the first few years and require 3 -5+ years to reach reproductive age. The University of

Minnesota Gladiolus breeding program developed certain gladiolus genotypes able to flower in the first year of seed sowing (Aljaser and Anderson, 2020; Anderson and Aljaser, 2020).

The designed probe for the UFC gene in Gladiolus is done by RAPiD genomics® LLC

(Gainesville, FL; http://rapid-genomics.com/home/) resulted in the generation in total of 433 sequence; 161 sequence among them had read hits of the UFC gene with various percentage of coverage. Of these, 34 selected sequences were chosen for this study, based on the largest length with two sequences per genotype due to presence of 2 alleles per genotype. These sequences represent the genomic sequence of UFC in gladiolus. The sequences were analyzed for gene prediction using the HMM-based gene structure prediction of FGENESH by having Arabidopsis thaliana (Generic) as the specific gene-finding parameter as the gene prediction is optimized for

Arabidopsis thaliana, the results confirmed the presence of UFC exons of a protein and coding sequence. The UFC coding sequences of each gladiolus genotype were analyzed in Genoeious® by pair-alignment with its genomic sequence to determine each exon. After the pair-alignment, the Coding sequence was then translated and aligned in multi-alignment process using MUSCLE alignment in the neighbor joining clustering method and CLUSTALW sequencing scheme with

UFC proteins of other species: Ananas comosus, Musa acuminata, Elaeis guineensis and

107

Asparagus officinalis, Arabidopsis thaliana and Glycine max. The UFC gene in Gladiolus

×hybridus was assigned to GhUFC as a label while genotype 14, G. dalenii ‘Carolina Primrose’, is assigned to GdUFC. In terms of the FLX gene, Gladiolus ×hybridus is assigned to GhFLX and genotype 14 G. dalenii ‘Carolina Primrose’ is GdFLX. There are two alleles of the UFC gene found in gladiolus. The designated alleles are A and B. Thus, the genes are designated as

GhUFC-A and GhUFC-B. The median number of amino acids of GhUFC-A protein is 420 amino acids across all gladiolus genotypes, with some genotypes having less than 420 amino acids and one genotype has 451 amino acids which could be due to an insertion, while GdUFC-A also has

420 amino acids. The second allele, GhUFC-B protein has a range of 375 to 410 amino acids,

GdUFC-B has 286 amino acids with an incomplete protein, missing many amino acids and a stop codon (Figure 2 and Figure 3). Genotype 9 and 12 have 100% amino acid sequences match in

GhUFC-A, while in GhUFC-B (GdUFC-B) genotype 14 and 8 have 100% match as well genotype 13 and 10, genotype 7 and 9, genotype 5 and 17 (data not shown). Gladiolus genotypes

1 and 15 were selected for pair-alignment with the species of comparison (Table 2), the table shows similarities of identity protein sequences and the percentage of GhUFC-A and GhUFC-B occur at range of ~30% to 57% across all species (Table 2). The intron-exon organization of

GhUFC-A 15 genotype is similar to Elaeis guineensis and Asparagus officinalis in term of exons splicing (Figure 4) while GhUFC-B genotype 1 has some similarity with Ananas comosus exons splicing. The remaining genotypes fall into these two configurations of the exon; the configuration shows the location of the conserved domain for UFC gene DUF966, which is found in Arabidopsis and other selected species of comparison. The DUF966 domain has the 92 amino acid and 93 amino acid of Ananas comosus. The multi alignment of the UFC protein conserved domain is overall conserved across species, although it is polymorphic (Figure 5). With the high identity matching in Gladiolus genotypes 1 and 15 exhibit a range of ~65% to ~86% across all

108

investigated species for the DUF966 domain of the UFC protein (Table 3). The GhUFC-A allele has a high identity across gladiolus genotypes, with the polymorphic exception of genotypes 3 and 7, due to (Figure 6). GhUFC-B is also conserved and identical in sequence with genotype 12, due to missing amino acids.

FLX gene was identified in gladiolus, the total genotypes which had presences of FLX are

12 out of 17 genotypes; 11 genotypes have GhFLX whereas GdFLX is identified in genotype 14

(Table 4). The range of amino acid proteins are from 146 to 254 amino acids missing the stop codon. Three genotype sequences (genotype 3, 6 and 16) have the longest amino acid chain. FLX is present in many species, in Arabidopsis alone it belongs to FLX gene family, FLX,

FLOWERING LOCUS C EXPRESSOR-LIKE 1 (FLL1), (FLL2), (FLL3) and (FLL4) (Lee &

Amasino 2013). Based on the pair-alignment, GhFLX and GdFLX matches FLL1 as high as 50% in amino acid identity (Table 4), the similarities of sequences in identity of gladiolus genotypes ranging from ~26% to ~65% across all investigated species of the whole FLX protein, the highest identity is in Ananas comosus match with ~65%. The multi-alignment for all FLX indicates that the tested gladiolus genotypes which has the longest amino acid sequences lacks exons (Figure

7), a pair-alignment test with Ananas comosus – FLX reveals GhFLX genotype 16 possibly lacks two exons and stop codon (Figure 8).

4.4 Discussion

The presence of a putative UFC gene in gladiolus is confirmed with two alleles, GhUFC-

A and GhUFC-B. It is highly possible that as the current gladiolus hybrids on the commercial market are mostly polyploids, primarily tetraploids with 60 chromosomes (2n = 4 × = 60)

(Bamford, 1935; Saito and Kusakari, 1972 and Ohri, 1985). Many of the gladiolus cultivars are 109

interspecific hybrids (Benschop 2010). Therefore, the presence of different alleles of genes is expected in the genotypes chosen for this study, as there are 13 genotypes are from University of

Minnesota gladiolus breeding program and the other 4 genotypes are from commercial gladiolus cultivars (Table 1). GhUFC-A gene has relatively ~50% identity with Musa acuminata, Elaeis guineensis and Asparagus officinalis (Table 2), the splicing of Elaeis guineensis and Asparagus officinalis exons is similar to GhUFC-A (Figure 4). GhUFC-B gene splicing is similar to Ananas comosus, Arabidopsis thaliana and Glycine max UFC gene splicing in the first 4 exons (Figure

4). These divergence in splicing of the UFC gene support the UFC gene being present in gladiolus with two alleles due to the possible of ploidy level, as many of gladiolus commercial are polyploids (Bamford, 1935). Further tests should be done to identify whether or not the UFC gene is also present in diploid gladiolus species, such as G. murielae, G. tristis and G. carneus since these three species only exist in the diploid form (Goldblatt & Manning 1998).

The UFC gene is responsive to vernalization by lowering expression alongside FLC and

DFC in Arabidopsis thaliana, as all these genes are in the cluster of vernalization stimulus region

(Finnegan et al 2004). FLC is a floral repressor, the overexpression of FLC results in a delay in flowering (Amasino 1999), while overexpression of UFC does not result in the altering flowering time (Finnegan et al 2004). Thus, UFC is adjacent to FLC both repressed by vernalization, yet

UFC does not show any influence in flowering time. This was observed herein because the genotypes for this study are a mixture of RGC-1, which are early flowering gladiolus able to reach flowering in the first year from seed and the classical later-flowering gladiolus which requires 3-5+ years to flower from seed. The multi-alignment of UFC protein of RGC-1 does not show any difference from non-RGC-1 genotypes, thus the UFC gene most likely isn’t involved in flowering, directly at least and this was proven in a UFC study in Arabidopsis thaliana (Sheldon

110

et al 2009). The main differences are shown in Figure 6 which represents the differences between alleles of UFC-A and UFC-B, regardless of the gladiolus genotypes tested herein (Figure 6).

In winter, vernalization suppress both FLC and UFC expression (Finnegan 2004), which allows floral genes integrators to promote flowering with the hypothesis would be that after vernalization and flowering, the UFC protein involvement is in embryogenesis and root initiation such that growth and branching occur in the spring season (since it doesn’t occur in the winter season). This could explain how both FLC and UFC are both negatively responsive to vernalization stimuli in the cluster genes area, while upstream of UFC is not responsive to vernalization (Finnegan et al 2004).

The identification of putative FLX gene in gladiolus raises the question whether gladiolus follows an Arabidopsis model of the flowering pathway. In the winter annual Arabidopsis thaliana, flowering is promoted after vernalization, which suppress the floral suppressor FLC which is upregulated by FRI, through activation of FRI complex of (FRI, FRL1, FRL2, FES1 and

SUF4) proteins in addition to FLX protein. FLX was proven to provide transcriptional activity for the FRI complex (Choi et al., 2011). A loss of function of FLX in Arabidopsis thaliana resulted in early flowering phenotypes (Dennis et al 2008), which indicates the clear role of FLX in flowering. The role of FLX in gladiolus has not been tested, particularly in RGC-1 genotypes and pedigrees; thus, FLX upregulation of FRI in gladiolus would be a rational approach. However,

FRI was not detected in gladiolus, using the primer design of Arabidopsis thaliana FRI

(At4g00650) because FRI gene was not detected in any monocotyledon species, thus the primer used is Arabidopsis thaliana, the test did not detect FRI gene in all 17 gladiolus genotypes without finding a single match (Aljaser, J. unpublished data). In addition, VRN2, the repressor of flowering in cereals and Arabidopsis thaliana was not detected in gladiolus, using the primer

111

design of Triticum monococum, Triticum durum and Hordeum vulgare (Aljaser, J. unpublished data, Appendix; A1, A2 and A3). This is not conclusive evidence as the genetic similarities between Arabidopsis thaliana and Gladiolus are low, given that GhUFC-A is ~32% and GhFLX is 50% identical to Arabidopsis thaliana genes. Therefore, there could be an FRI gene in gladiolus but would require better primer design to locate the gene, because the presence of the

GhFLX gene might indicate in the presence of other flowering repressor genes as FRI protein upregulates FLX in Arabidopsis thaliana and this is part of the flowering pathway (Choi et al.,

2011). In addition, Musa acuminata, Elaeis guineensis and Ananas comosus are all monocots and tropical species which have FLX and SUF4 genes which are part of FRI complex (Amasino and

Michaels 2010). This indicates the presence of some of FRI complex components while a lack of identification of FRI gene itself creates divergent possibilities: either there are FRI and FLC genes in these species or a lack of these genes and the presence of SUF4 and FLX genes have other unknown flowering pathway purposes. Since GhFLX and GdFLX have similarities to FLL1, reaching up to 50% identity in amino acids, FLX gene is part of gene family, FLX-LIKE1

(FLL1), FLX-LIKE2 (FLL)), FLX-LIKE3 (FLL3), FLX-LIKE4 (FLL4) (Choi et al 2011; Lee et al

2013). While the role of FLL1 in flowering pathway is not proven, FLX and FLL4 are most crucial genes in control of flowering time in Arabidopsis (Lee and Amasino, 2013).

The relation between UFC and FLX is addressed in Dennis et al 2008, pointing that a mutation in FLX would influence in UFC expression, in which UFC expression is reduced in flx mutant in Arabidopsis thaliana, this was a clear study to establish the influence of between the two genes.

The next step in this research would be to identify the UFC gene in diploid gladiolus species to determine if the allele is similar to GhUFC-A, GhUFC-B or a third different allele,

112

using diploid species would simplify the study to determine the function of UFC protein in gladiolus by silencing and knocking out the gene. Locating the physical location of the UFC gene in Gladiolus will help in testing if there are other UFC genes in gladiolus as part of a UFC gene family, since the first discovered UFC gene (At5g10150) in Arabidopsis thaliana is located in the cluster genes UFC, FLC and DFC on chromosome 5 (Finnegan et al 2004). UFC is also designated SOK2 and the other UFC genes are grouped in SOK gene family such as SOK1

(At1g05577), SOK3 (At2g28150), SOK4 (At3g46110) and SOK5 (At5g59790) (Yoshida and

Weijers et al 2019).

In terms of the FLX gene, identifying the FLX gene in Eurasian species of Gladiolus, particularly G. italicus, G. imbricatus and G. communis, would be informative since these perennial species live in temperate habitats that require vernalization to break corm dormancy in winter season (Cohen and Barzilay, 1991; Cohat, 1993; Gonzalez, 1998). Conversely, identifying

FLX in subtropical gladiolus species, such as G. crassifolius, G. laxiflorus and G. atropurpureus

(Goldblatt 1996), would allow the comparison of FLX among these different habitats may support the influence in FLX in the process of flowering. Furthermore, the use of transgene silencing of

FLX in gladiolus would determine whether or not FLX influences the production of a rapid flowering phenotype gladiolus (RGC-1), as was reported in loss of function of flx in Arabidopsis thaliana (Dennis et al 2008). In conclusion, the discovery of UFC and FLX genes in gladiolus provides insight of the better understanding for flowering and vernalization response in ornamental geophytes.

113

4.5 Literature Cited

Aljaser, J.A. (2020). Gladiolus breeding for rapid generation cycling for potted production and

the discovery of gladiolus genes, UFC and FLX. University of Minnesota, Minneapolis,

U.S.A. Ph.D. Dissertation.

Andersson, C. R., Helliwell, C. A., Bagnall, D. J., Hughes, T. P., Finnegan, E. J., Peacock, W. J.,

& Dennis, E. S. (2008). The FLX gene of Arabidopsis is required for FRI-dependent

activation of FLC expression. Plant and cell physiology, 49(2), 191-200.

Anderson NO, Carter J, Hershman A, Houseright V (2015) Rapid generation cycling enhances

selection rate of gladiolus xhybridus. Acta Hort. 1087:429-435

Anderson, N. O. (2019). Selection tools for reducing generation time of geophytic herbaceous

perennials. In XIII International Symposium on Flower Bulbs and Herbaceous Perennials

1237 (pp. 53-66).

Amasino, R. M., & Michaels, S. D. (2010). The timing of flowering. Plant Physiology, 154(2),

516-520.

Bamford, R., 1935. The chromosome number in Gladiolus. Jour. Agr. Res. 51, pp.945-950.

Benschop, M., Kamenetsky, R., Le Nard, M., Okubo, H., & De Hertogh, A. (2010). 1 The Global

Flower Bulb Industry: Production, Utilization, Research. Horticultural Reviews, 36(1), 1-115.

Blázquez, M. A., & Weigel, D. (2000). Integration of floral inductive signals in

Arabidopsis. Nature, 404(6780), 889-892.

114

Castro, M., Castro, S., Figueiredo, A., Husband, B., & Loureiro, J. (2018). Complex

cytogeographical patterns reveal a dynamic tetraploid–octoploid contact zone. AoB

Plants, 10(2), ply012. Available from:

https://academic.oup.com/aobpla/article/10/2/ply012/4857208

Chengke, L., and Lei, T. (2017) Bioinformatics analysis of rice DUF966 gene family. Molecular

Plant Breeding, 15(12), 4791-4796.

Choi, K., Kim, J., Hwang, H. J., Kim, S., Park, C., Kim, S. Y., & Lee, I. (2011). The FRIGIDA

complex activates transcription of FLC, a strong flowering repressor in Arabidopsis, by

recruiting chromatin modification factors. The Plant Cell, 23(1), 289-303.

Cohen, A., & Barzilay, A. (1991). Miniature gladiolus cultivars bred for winter

flowering. HortScience, 26(2), 216-218.

Cohat, J. (1993). Gladiolus. W: The physiology of flower bulbs. Red. De Hertogh, A. & Le

Nard, M. Elsevier, Amsterdam, p. 297-320. de Oliveira, R. R., Cesarino, I., Mazzafera, P., & Dornelas, M. C. (2014). Flower development in

Coffea arabica L.: new insights into MADS-box genes. Plant reproduction, 27(2), 79-94.

Ding, L., Kim, S. Y., & Michaels, S. D. (2013). FLOWERING LOCUS C EXPRESSOR family

proteins regulate FLOWERING LOCUS C expression in both winter-annual and rapid-cycling

Arabidopsis. Plant physiology, 163(1), 243-252.

Droc, G., Lariviere, D., Guignon, V., Yahiaoui, N., This, D., Garsmeur, O., Dereeper, A.,

Hamelin, C., Argout, X., Dufayard, J.-F., Lengelle, J., Baurens, F.-C., Cenci, A., Pitollat, B.,

D'Hont, A., Ruiz, M., Rouard, M., Bocs, S. The Banana Genome

Hub. Database (2013) doi:10.1093/database/bat035 115

Ehrich, L. (2013). Flowering in South African Iridaceae. (eds) Ramawat, K. G., & Merillon, J.

M. Bulbous Plants: Biotechnology. P. 248-269.

Fadón, E., Herrero, M., & Rodrigo, J. (2015). Flower bud dormancy in Prunus species.

In Advances in Plant Dormancy (pp. 123-135). Springer, Cham.

Finnegan, E. J., Sheldon, C. C., Jardinaud, F., Peacock, W. J., & Dennis, E. S. (2004). A cluster

of Arabidopsis genes with a coordinate response to an environmental stimulus. Current

Biology, 14(10), 911-916.

Grant, D., Nelson, R.T., Cannon, S.B. and Shoemaker, R.C. (2010) SoyBase, the USDA-ARS

soybean genetics and genomics database.

Nucl. Acids Res. (2010) 38 (suppl 1): D843-D846. doi: 10.1093/nar/gkp798.

https://www.soybase.org/

Goldblatt, P. (1996). Gladiolus in tropical Africa. Systematics, biology & evolution. Timber

Press, Inc..

Goldblatt, P., & Manning, J. (1998). Gladiolus in southern Africa. Fernwood Press (Pty) Ltd..

Gonzalez, A., Banon, S., Fernandez, J. A., Franco, J. A., Casas, J. L., & Ochoa, J. (1998).

Flowering responses of Gladiolus tristis (L.) after exposing corms to cold treatment. Scientia

horticulturae, 74(4), 279-284.

Guo, X, Yu, C, Luo, L, Wan, H, Zhen, N, Xu, T, Tan, J, Pan, H, Zhang, Q. (2017). Transcriptome

of the floral transition in Rosa chinensis ‘Old Blush’. BMC genomics, 18(1), 199.

Harkess, A., Zhou, J., Xu, C., Bowers, J. E., Van der Hulst, R., Ayyampalayam, S., Mercati, F.,

Riccardi, P., McKain, M. R., Kakrana, A., Tang, H., Ray, J., Groenendijk, J., Arikit, S.,

116

Mathioni, S. M., Nakano, M., Shan, H., Telgmann-Rauber, A., Kanno, A., Yue, Z., Chen, H.,

Li, W., Chen, Y., Xu, X., Zhang, Y., Luo, S., Chen, H., Gao, J., Mao, Z., Piras, J. C., Luo,

M., Kudrna, D., Wing, R. A., Meyers, B. C., Yi, K., Kong, H., Lavrijsen, P., Sunseri, F.,

Falavigna, A., Ye, Y., Leebens-Mack, J. H., Chen, G. (2017). The asparagus genome sheds

light on the origin and evolution of a young Y chromosome. Nature communications, 8(1), 1-

10. http://asparagus.uga.edu/

Kamenetsky R, Zaccai M, Flaishman MA (2012) Florogenesis. In: Kamenetsky R, Okubo H

(eds) Ornamental geophytes: from basic science to sustainable production. CRC press,

Florida, pp 197-232

Kamo, K. K., Krens, F. A., & Ziv, M. (2012). Biotechnology for the modification of horticultural

traits in geophytes. Ornamental Geophytes; from basic science to sustainable production. R

Kamenetsky & H Okubo (eds.), CRC Press, Taylor & Francis Group, Boca Raton, 159-195.

Kelley, L. A., Mezulis, S., Yates, C. M., Wass, M. N., & Sternberg, M. J. (2015). The Phyre2

web portal for protein modeling, prediction and analysis. Nature protocols, 10(6), 845-858.

http://www.sbg.bio.ic.ac.uk/~phyre2/html/page.cgi?id=index

Kumar, G., Arya, P., Gupta, K., Randhawa, V., Acharya, V., & Singh, A. K. (2016). Comparative

phylogenetic analysis and transcriptional profiling of MADS-box gene family identified

DAM and FLC-like genes in apple (Malus x domestica). Scientific reports, 6, 20695.

Ohri, D., 1985. Cytogenetics of Garden Gladiolus. Cytologia, 50(2), pp.213-231.

Lee, J., & Amasino, R. M. (2013). Two FLX family members are non-redundantly required to

establish the vernalization requirement in Arabidopsis. Nat. Commun., 4, 2186. doi:

10.1038/ncomms3186

117

Lee, J. H., Yoo, S. J., Park, S. H., Hwang, I., Lee, J. S., & Ahn, J. H. (2007). Role of SVP in the

control of flowering time by ambient temperature in Arabidopsis. Genes &

development, 21(4), 397-402.

Levy, Y. Y., Mesnage, S., Mylne, J. S., Gendall, A. R., & Dean, C. (2002). Multiple roles of

Arabidopsis VRN1 in vernalization and flowering time control. Science, 297(5579), 243-246.

Li, Y. F., Zhang, M. F., Zhang, M., & Jia, G. X. (2017). Analysis of global gene expression

profiles during the flowering initiation process of Lilium × formolongi. Plant molecular

biology, 94(4-5), 361-379.

Luo, X., Lu, H., Yuan, L., Jia, Y., Wu, Y., & Deng, Q. (2016). Cloning and expression analysis of

gibberellin receptor gene in Gladiolus hybridus. Acta Botanica Boreali-Occidentalia

Sinica, 36(11), 2152-2158.

Michaels, S.D. and Amasino, R.M. (1999) FLOWERING LOCUS C encodes a novel MADS

domain protein that acts as a repressor of flowering. Plant Cell 11, 949–956

Michaels, S. D., Himelblau, E., Kim, S. Y., Schomburg, F. M., & Amasino, R. M. (2005).

Integration of flowering signals in winter-annual Arabidopsis. Plant Physiology, 137(1), 149-

156.

Monteagudo, A., Igartua, E., Contreras-Moreira, B., Gracia, M. P., Ramos, J., Karsai, I., & Casas,

A. M. (2019). Fine-tuning of the flowering time control in winter barley: the importance of

HvOS2 and HvVRN2 in non-inductive conditions. BMC plant biology, 19(1), 1-14.

Moon, J., Suh, S. S., Lee, H., Choi, K. R., Hong, C. B., Paek, N. C., Kim, S. G., & Lee, I. (2003).

The SOC1 MADS‐box gene integrates vernalization and gibberellin signals for flowering in

Arabidopsis. The Plant Journal, 35(5), 613-623. 118

National Center for Biotechnology Information (2016a). Accession No. XM_009383889.2,

PREDICTED: Musa acuminata subsp. malaccensis protein UPSTREAM OF FLC

(LOC103970210), transcript variant X1, mRNA; [cited 2020 Jul 6]. Available from:

https://www.ncbi.nlm.nih.gov/nuccore/XM_009383889

National Center for Biotechnology Information (2016b). Accession No. XM_008801571.2,

PREDICTED: protein FLOWERING LOCUC C EXPRESSOR-like [Phoenix dactylifera];

[cited 2020 Jul 6]. Available from: https://www.ncbi.nlm.nih.gov/nuccore/XM_008801571.2

National Center for Biotechnology Information (2016c). Accession No. XP_009420070.1,

PREDICTED: protein FLX-like 1 [Musa acuminata subsp. malaccensis]; [cited 2020 Jul 6].

Available from: https://www.ncbi.nlm.nih.gov/protein/695063148

National Center for Biotechnology Information (2017a). Accession No. XM_010920607.2,

PREDICTED: protein UPSTREAM OF FLC [Elaeis guineensis]; [cited 2020 Jul 6].

Available from: https://www.ncbi.nlm.nih.gov/nuccore/XM_010920607.2

National Center for Biotechnology Information (2017b). Accession No. XM_010924316.2,

protein FLC EXPRESSOR [Elaeis guineensis]; [cited 2020 Jul 6]. Available from:

https://www.ncbi.nlm.nih.gov/nuccore/XM_010924316.2

National Center for Biotechnology Information (2017c). Accession No. XP_020095672.1, protein

FLX-like 1 [Ananas comosus]; [cited 2020 Jul 6]. Available from:

https://www.ncbi.nlm.nih.gov/protein/XP_020095672.1

National Center for Biotechnology Information (2019a). Accession No. XP_010922618.1, protein

FLC EXPRESSOR [Elaeis guineensis]; [cited 2020 Jul 6]. Available from:

https://www.ncbi.nlm.nih.gov/protein/XP_010922618.1

119

National Center for Biotechnology Information (2019b). Accession No. NP_001154541.1,

protein FLC EXPRESSOR [Arabidopsis thaliana]; [cited 2020 Jul 6]. Available from:

https://www.ncbi.nlm.nih.gov/protein/NP_001154541.1

National Center for Biotechnology Information (2019c). Accession No. NP_566492.1, structural

maintenance of chromosomes domain protein [Arabidopsis thaliana]; [cited 2020 Jul 6].

Available from: https://www.ncbi.nlm.nih.gov/protein/NP_566492.1

National Center for Biotechnology Information (2019d). Accession No. NP_001320766.1,

sarcolemmal membrane-associated protein [Arabidopsis thaliana]; [cited 2020 Jul 6].

Available from: https://www.ncbi.nlm.nih.gov/protein/NP_001320766.1

National Center for Biotechnology Information (2019e). Accession No. NP_564678.1, DNA

double-strand break repair protein [Arabidopsis thaliana]; [cited 2020 Jul 6]. Available from:

https://www.ncbi.nlm.nih.gov/protein/NP_564678.1

National Center for Biotechnology Information (2019f). Accession No. NP_001119474.1, FLX-

like protein [Arabidopsis thaliana]; [cited 2020 Jul 6]. Available from:

https://www.ncbi.nlm.nih.gov/protein/NP_001119474.1

Noy-Porat, T., Flaishman, M. A., Eshel, A., Sandler-Ziv, D., & Kamenetsky, R. (2009).

Florogenesis of the Mediterranean geophyte Narcissus tazetta and temperature requirements

for flower initiation and differentiation. Scientia horticulturae, 120(1), 138-142.

Noy-Porat, T., Kamenetsky, R., Eshel, A., & Flaishman, M. A. (2010). Temporal and spatial

expression patterns of the LEAFY homologue NLF during florogenesis in Narcissus

tazetta. Plant Science, 178(2), 105-113.

120

Oldhousegardens. (2020). Gladiolus: Lost Forever?. Available from:

https://oldhousegardens.com/display/category/BackSoonGladiolus?bulb=SGL30

Rotem, N., Shemesh, E., Peretz, Y., Akad, F., Edelbaum, O., Rabinowitch, H. D., Sela, I., &

Kamenetsky, R. (2007). Reproductive development and phenotypic differences in garlic are

associated with expression and splicing of LEAFY homologue gaLFY. Journal of

Experimental Botany, 1133-1141.

Ruelens, P., De Maagd, R. A., Proost, S., Theißen, G., Geuten, K., & Kaufmann, K. (2013).

FLOWERING LOCUS C in monocots and the tandem origin of angiosperm-specific MADS-

box genes. Nature Communications, 4(1), 1-8.

Rotem, N., David-Schwartz, R., Peretz, Y., Sela, I., Rabinowitch, H. D., Flaishman, M., &

Kamenetsky, R. (2011). Flower development in garlic: the ups and downs of gaLFY

expression. Planta, 233(5), 1063-1072.

Turner, A. S., Faure, S., Zhang, Y., & Laurie, D. A. (2013). The effect of day-neutral mutations

in barley and wheat on the interaction between photoperiod and vernalization. Theoretical

and applied genetics, 126(9), 2267-2277.

Saito, K. and Kusakari, K., 1972. Studies on the Occurrence of Polyploidy and its Contribution to

the Flower Plants Breeding.: IX. Cytological Observations on the Mechanism of Decreased

Fertility in the Summer-flowering Tetraploid Cultivars of Gladiolus grandiflorus

Hort. Japanese Journal of Breeding, 22(2), pp.75-82.

Sanusi, N. S. N. M., Rosli, R., Halim, M. A. A., Chan, K. L., Nagappan, J., Azizi, N., Amiruddin,

N., Tatarinova, T. V., & Low, E. T. L. (2018). PalmXplore: oil palm gene

database. Database, 2018. http://palmxplore.mpob.gov.my

121

Sheldon, C. C., Jean Finnegan, E., James Peacock, W., & Dennis, E. S. (2009). Mechanisms of

gene repression by vernalization in Arabidopsis. The Plant Journal, 59(3), 488-498.

Simpson, G. G. (2004). The autonomous pathway: epigenetic and post-transcriptional gene

regulation in the control of Arabidopsis flowering time. Current opinion in plant

biology, 7(5), 570-574.

Simpson, G. G., & Dean, C. (2002). Arabidopsis, the Rosetta stone of flowering

time?. Science, 296(5566), 285-289.

Srikanth, A., & Schmid, M. (2011). Regulation of flowering time: all roads lead to

Rome. Cellular and molecular life sciences, 68(12), 2013-2037.

Tadege, M., Sheldon, C. C., Helliwell, C. A., Stoutjesdijk, P., Dennis, E. S., & Peacock, W. J.

(2001). Control of flowering time by FLC orthologues in Brassica napus. The Plant

Journal, 28(5), 545-553.

Taylor, A. (2009). Functional genomics of photoperiodic bulb initiation in onion (Allium

cepa) (Doctoral dissertation, University of Warwick).

Taylor, A., Massiah, A. J., & Thomas, B. (2010). Conservation of Arabidopsis thaliana

photoperiodic flowering time genes in onion (Allium cepa L.). Plant and cell

physiology, 51(10), 1638-1647.

The Arabidopsis Information Resource (TAIR),

[https://www.arabidopsis.org/servlets/TairObject?id=136005&type=locus], on

www.arabidopsis.org, [May 2020].

122

Wang, A. J., Tang, J. F., Zhao, X. Y., & Zhu, L. H. (2008). Isolation of LiLFY1 and its expression

in lily (Lilium longiflorum Thunb.). Agricultural Sciences in China, 7(9), 1077-1083.

Wayside Gardens. (2020). Gladiolus Glamini® Mixed. Available from:

https://www.waysidegardens.com/glamini-mixed-gladiolus/p/27192-PK-10/

Xu, H., Yu, Q., Shi, Y., Hua, X., Tang, H., Yang, L., Ming, R., & Zhang, J. (2018). PGD:

pineapple genomics database. Horticulture research, 5(1), 1-9. Pineapple Genomics Database

http://pineapple.angiosperms.org/pineapple/html/index.html

Yan, L., Loukoianov, A., Blechl, A., Tranquilli, G., Ramakrishna, W., SanMiguel, P., Bennetzen,

J., Echenique, V. & Dubcovsky, J. (2004). The wheat VRN2 gene is a flowering repressor

down-regulated by vernalization. Science, 303(5664), 1640-1644.

Ye, J., Zhong, T., Zhang, D., Ma, C., Wang, L., Yao, L., Zhang, Q., Zhu, M., & Xu, M. (2019).

The auxin-regulated protein ZmAuxRP1 coordinates the balance between root growth and

stalk rot disease resistance in maize. Molecular plant, 12(3), 360-373.

Yoshida, S., Van Der Schuren, A., Van Dop, M., Van Galen, L., Saiga, S., Adibi, M., Möller, B.,

Ten Hove, C. A., Marhavy, P., Smith, R., Friml, J., and Weijers D. (2019). A SOSEKI-based

coordinate system interprets global polarity cues in Arabidopsis. Nat. plants, 5(2), 160-166.

doi: 10.1038/s41477-019-0363-6

Zhao, T., Ni, Z., Dai, Y., Yao, Y., Nie, X., & Sun, Q. (2006). Characterization and expression of

42 MADS-box genes in wheat (Triticum aestivum L.). Molecular Genetics and

Genomics, 276(4), 334.

Zhu, Q. H., & Helliwell, C. A. (2011). Regulation of flowering time and floral patterning by

miR172. Journal of experimental botany, 62(2), 487-495. 123

Zimin, A. V., Marçais, G., Puiu, D., Roberts, M., Salzberg, S. L., & Yorke, J. A. (2013). The

MaSuRCA genome assembler. Bioinformatics, 29(21), 2669-2677.

124

Table 1. The Gladiolus genotypes used in this study, their codes and whether they are Rapid

Generation Cycling (RGC): + is for RGC genotypes and – for Non-RGC genotypes. All gladiolus genotypes were tested for the presence of UFC gene, FLX gene and FRI gene.

Genotype Code RGC Gladiolus ×hybridus 21213 1 + Gladiolus ×hybridus 2220 2 + Gladiolus ×hybridus 2231 3 + Gladiolus ×hybridus 2337 4 + Gladiolus ×hybridus 35314 5 + Gladiolus ×hybridus 3923 6 + Gladiolus ×hybridus 3931 7 + Gladiolus ×hybridus 74210 8 + Gladiolus ×hybridus 7736 9 + Gladiolus ×hybridus 28236 10 - Gladiolus ×hybridus 15531 11 - Gladiolus ×hybridus 20732 12 - Gladiolus ×hybridus 60314 13 - Gladiolus dalenii ‘Carolina Primrose’ 14 - Gladiolus ×hybridus ‘Beatrice’ 15 - Gladiolus ×hybridus ‘Glamini’® 16 - Gladiolus ×hybridus ‘Vista 17 -

125

Table 2. The identity of amino acid sequences and number (%) of two UFC proteins (GhUFC-A, GhUFC-B) in two Gladiolus (genotypes 1 and

15) in relation to other species (Gene locus/ID) through pair alignment. The similarities of sequences in identity of gladiolus genotypes ranges from ~30% to 57% across all investigated species for the whole UFC protein. The alignment is done with MUSCLE alignment using the neighbor joining clustering method and the CLUSTALW sequencing scheme (Geneious®).

Gladiolus ×hybridus genotype

Genotype 1 Genotype 15

Species (Gene locus/ID) GhUFC-A (%) GhUFC-B (%) GhUFC-A (%) GhUFC-B (%)

Ananas comosus (Aco009327) 189 (33.51%) 249 (53.21%) 191 (34.35%) 231 (52.98%)

Musa acuminata (GSMUA_Achr5T28540_001) 251 (54.09%) 170 (39.35%) 250 (57.74%) 148 (31.36%)

Elaeis guineensis (p5.00_sc00099_p0095) 254 (52.05%) 172 (40.86%) 250 (51.23%) 155 (39.85%)

Asparagus officinalis (evm.model.AsparagusV1_08.3493) 213 (46.61%) 148 (35.58%) 213 (50.0%) 135 (29.87%)

Arabidopsis thaliana (AT5G10150) 137 (28.54%) 129 (29.79%) 138 (31.72%) 117 (29.32%)

Glycine max (Glyma.11G193000.1) 181 (33.64%) 226 (52.19%) 188 (35.67%) 207 (52.01%)

126

Table 3. Identity of amino acid sequences, number (%) of UFC proteins in the conserved domain DUF966 in Gladiolus genotypes 1 and 15 in relation to other species through pair alignment. Gladiolus genotypes 1 and 15 exhibit a range of ~65% to ~86% across all investigated species for the DUF966 domain of UFC protein.

Gladiolus ×hybridus genotype

Genotype 1 Genotype 15 Species (Gene locus/ID) GhUFC-A (%) GhUFC-B (%) GhUFC-A (%) GhUFC-B (%) Ananas comosus (Aco009327) 70 (76.09%) 79 (84.95%) 79 (84.95%) 79 (84.95%)

Musa acuminata (GSMUA_Achr5T28540_001) 78 (84.78%) 71 (78.02%) 78 (84.78%) 71 (78.02%)

Elaeis guineensis (p5.00_sc00099_p0095) 79 (85.87%) 72 (79.12%) 79 (85.87%) 72 (79.12%)

Asparagus officinalis (evm.model.AsparagusV1_08.3493) 75 (82.42%) 70 (76.09%) 75 (82.42%) 70 (76.09%)

Arabidopsis thaliana (AT5G10150) 61 (66.30%) 61 (64.89%) 62 (67.39%) 61 (64.89%)

Glycine max (Glyma.11G193000.1) 70 (76.92%) 75 (82.42%) 70 (76.92%) 75 (81.52%)

127

Table 4. Number (%) of amino acid sequences of GhFLX protein in Gladiolus genotypes 3 and 6 (genotype 16 is identical to genotype 6) in relation to the other species through pair alignment. The similarities of sequences in identity of gladiolus genotypes ranged from ~26% to ~65% across all investigated species of the whole FLX protein. The alignment is done in MUSCLE, using the neighbor joining clustering method and the

CLUSTALW sequencing scheme (Geneious®).

Gladiolus ×hybridus genotypes

Species (Accession no.) GhFLX Genotype 3 (%) GhFLX Genotype 6 & 16 (%)

Ananas comosus (XP_020095672.1) 180 (64.98%) 177 (64.60%)

Musa acuminata (XP_009420070.1) 122 (48.03%) 122 (48.03%)

Elaeis guineensis (XP_010922618.1) 132 (50.00%) 131 (49.62%)

Arabidopsis thaliana – FLX (NP_001154541.1) 82 (30.48%) 82 (30.48%)

Arabidopsis thaliana – FLL1 (NP_566492.1) 135 (50.00%) 135 (50.00%)

Arabidopsis thaliana – FLL2 (NP_001320766.1) 96 (36.09%) 94 (35.34%)

Arabidopsis thaliana – FLL3 (NP_564678.1) 104 (34.67%) 104 (34.67%)

Arabidopsis thaliana – FLL4 (NP_001119474.1) 67 (26.38%) 68 (26.77%)

Glycine max (Glyma.15g269300) 156 (57.14%) 155 (56.78%)

128

Figure 1. Model represents portion of flowering pathway regarding the role of FLX gene in flowering along with UFC gene in Arabidopsis and temperate dicots. FLX upregulates FRI which also upregulates the expression of the FLC protein and suppresses flowering by repressing the expression of the floral integrator SOC1 and FT. Vernalization pathway downregulates the expression of FLX, FRI and FLC genes allowing the floral integrators to initiate flowering in vegetative state of dicots, while the vernalization pathway also downregulates the UFC gene

(Finnegan, 2004). In monocots, FLX, FLC,SOC1 and FT have been identified (Noy-Porat 2009;

Amasino and Michaels 2010; Jia et al 2017; Monteagudo, 2019), while UFC and FLX have just been identified in gladiolus (in the current experiments). However, FRI was not identified either by lacking the presence of these repressor genes or monocots relying on other options of the flowering pathway genes.

129

Figure 2. Multi-alignment of UFC coding sequence in Gladiolus ×hybridus (GhUFC) and Gladiolus dalenii (GdUFC), the alignment is for the 17 genotypes, each genotype has 2 alleles, allele A and allele B: GhUFC-A, GdUFC-A, GhUFC-B, GdUFC-B. Both alleles has 4 exons but allele A size is larger in coding sequence than allele B. The alignemtn shows insertion and missing coding sequences in some genotypes. The multi- alignment is done in MUSCLE pair-alignment using neighbor joining cluster method and CLUSTALW sequencing scheme (Geneious)®

130

Figure 3. Multi-alignment of UFC amino acid sequence in Gladiolus ×hybridus (GhUFC) and Gladiolus dalenii (GdUFC), the alignment is for the 17 genotypes, each genotype has 2 alleles, allele A and allele B: GhUFC-A, GdUFC-A, GhUFC-B, GdUFC-B. Both alleles has 4 exons but allele A size is larger in amino acid sequence than allele B. The alignemtn shows insertion and missing amino acid sequences in some genotypes.

The alighnment identify conserved amino acid sequences (green color). The multi-alignment is done in MUSCLE pair-alignment using neighbor joining cluster method and CLUSTALW sequencing scheme (Geneious)® 131

Figure 4. Intron-exon configuration of the UFC genes in Gladiolus ×hybridus of genotypes 1 and 15 in relation to several species. Monocot species are highlighted in blue: Ananas comosus, Musa acuminata, Elaeis guineensis and Asparagus officinalis. Dicot species highlighted in green 132

for Arabidopsis thaliana and Glycine max. The intron-exon organization of GhUFC-A genotype 15 is similar to Elaeis guineensis and Asparagus officinalis in terms of exons splicing while GhUFC-B genotype 1 has some similarity with Ananas comosus exons. Sequences were aligned based on first exon sequences. Total length of the gene’s coding region is listed on the right of each respected species. The red line represents the conserved domain DUF966 in relation to the location of the domain with exon configuration.

133

Figure 5. Alignment of the globular region containing DUF966 domain of UFC proteins from Gladiolus ×hybridus of genotypes 1 and 15, Ananas comosus, Musa acuminata, Elaeis guineensis Asparagus officinalis, Arabidopsis thaliana and Glycine max. Green coloration shows identical amino acid sequence; yellow color highlights the polymorphisms while red color shows the cytosine amino acid. The conserved domain DUF966 is 92 amino acids.

134

Figure 6. The phylogenetic tree of all UFC genotypes in Gladiolus; genetic distances were computed using the Tamura-Nei method and are in the units of the number of base substitutions per site. The tree build using the Neighbor-Joining method and the bootstrap test was performed for each tree (500 replicates) and the tree format is organized and ordered with a scale bar of 0.2 (Geneious®) 135

Figure 7. Multi-alignment of FLX amino acid sequence in Gladiolus ×hybridus (GhFLX) with other species; Arabidopsis thaliana, Ananas comosus, Elaeis guineensis, Musa acuminata and Glycine max. The alignment is for the 3 gladiolus genotypes (3, 6 and 16) each genotype has 3 exons. The alignment shows missing amino acid sequences in gladiolus genotypes 3, 6 and 16 as amino acid sequences does not have a stop codon. The alignment identifies conserved amino acid sequences (green color). Note Arabidopsis thaliana – FLL4 is a functional protein which has two exons only (Lee and Amasino, 2013). The multi-alignment is done in MUSCLE pair-alignment using neighbor joining cluster method and CLUSTALW sequencing scheme (Geneious)® 136

Figure 8. Pair alignment of FLX protein in Ananas comosus and Gladiolus ×hybridus genotype 16 showing the complete FLX protein in Ananas comosus with five exons while incomplete FLX protein in Gladiolus ×hybridus (GhFLX) which lacks the remaining two exons and stop codon.

The alignment is done in MUSCLE pair-alignment using neighbor joining cluster method and CLUSTALW sequencing scheme (Geneious)®

137

5.0 Conclusion

Rapid generation cycling is a powerful tool can be implemented to reduce the juvenility period in perennial crop and should be applied in breeding program ideotype. Although it is possible to annualize a perennial crop through genetically modifying the flowering pathway by overexpression a positive flowering regulator or inserting blocker of flowering suppressor

(Srinivasan and Dardick et al 2012), yet those methods are biotechnological methods and require much regulation and approval, or sometimes rejected. Therefore, conventional breeding methods for early flowering are wildly accepted and does not require much regulation for releasing the cultivar.

Geophytes such as gladiolus are perennial and take 3 – 5 years to flower from seed; likewise, tulip and daffodil each take 4 – 6 years (Fortanier, 1973). The University of Minnesota gladiolus breeding program aims to reduce the number of years to reach flowering (Anderson et al., 2015) and successfully “annualized” the perennial gladiolus to flower from seed in the first year. Gladiolus breeding for rapid generation cycling was first accomplished by Roemer

(Roemer, 1907) and was commercialized in market (Burpee®, 1913), Minnesota Gladiolus

Breeding Program successfully breed for rapid generation cycling using different genotypes than

Roemer. The RGC lines was superior than commercial cultivars, by reaching to flowering from seed in less than a year, petite height suitable for potted production without the need of using plant growth retardants to reduce the height for seed propagated gladiolus.

The use of plant growth retardants in floriculture crops is demanding to create uniformity, growth retardants are used regularly on potted production (Miller et al, 2012), the taller the crops is the higher the concentration used. Gladiolus is relatively tall flowering crops ranging 1-2 meters in height, reducing the height using plant growth retardants require higher concentration.

138

However, high concentrations show inhibition to flowering as gibberellins is one of the factors of flowering (Ehrich, 2013; Kamenetsky, et al., 2012). corms of RGC genotypes able to reach flowering under the use of high GA-inhibitor concentration. In addition, RGC genotypes able to reach flowering under such inhibiting factors while maintaining a shorter stature proves the superiority of RGC lines to be candidate for potted production gladiolus. Another trait in RGC genotypes is their ability to re-sprout by having reduce dormancy, which break the typical requirement of vernalization in gladiolus.

Vernalization is the exposure of cold weather for time period then subsequent induction of flowering. Temperate crops require vernalization to flower (Dubcovsky, 2009). The lack of vernalization requirement in RGC genotypes can acts as a mutant in the studying of vernalization genes in gladiolus. These features add value of RGC genotypes from economic value and research value as well. In ornamental plants, flowering is a crucial step for the success of cut flower production. Therefore, understanding the flowering pathway and the gene expression is important for efficient selective breeding for rapid generation cycling. Furthermore, the identification of flowering genes in geophytes is poorly understood. An important gene in flowering is FLOWERING LOCUS C (FLC) which is a major flowering repressor found in

Arabidopsis and many dicot species, FLC plays vital role in the control of flower initiation

(Michaels and Amasino 1999). However, the lack of identifying FLC in many monocot species and discovery of another mechanism of flowering in monocots, such as plant age in both maize and rice (Leeggangers et al., 2013), led researchers to suggest that monocot geophytes could also be FLC-independent for flower initiation. The search for FLC gene and its regulatory genes in gladiolus is a step to uncover the flowering pathway in geophytes. To uncover if FLC is present in Gladiolus, we searched for linked genes with FLC. In Arabidopsis, FLC is adjacent to two genes, UPSTREAM OF FLOWERING LOCUS C (UFC) and DOWNSTREAM OF FLOWERING 139

LOCUS C (DFC). The both UFC and FLC genes are downregulated by vernalization FLC

(Dennis et al., 2004). The discovery of UFC in gladiolus as well FLOWERING LOCUS C

EXPRESSOR (FLX) is crucial to establish the flowering pathway, UFC gene being adjacent to

FLC in Arabidopsis and responses to vernalization and FLX a gene that is upregulates FRIGIDA

(FRI) which upregulates FLC is an early indicator in the presence of FLC homologue in gladiolus.

All these finds are important to understand flowering mechanism and gene of flowering pathway to utilize the knowledge in order to produce potted gladiolus either by using seed as germplasm or corm of RGC genotypes (Appendix; A4).

140

5.1 Literature cited

Anderson, N.O., J. Carter, A. Hershman, and V. Houseright. 2015. Rapid generation cycling

enhances selection rate of Gladiolus xhybridus. Acta Hort. 1087:429-435.

Distelfeld, A., Li, C., & Dubcovsky, J. (2009). Regulation of flowering in temperate

cereals. Current opinion in plant biology, 12(2), 178-184.

Ehrich, L. (2013). Flowering in South African Iridaceae. (eds) Ramawat, K. G., & Merillon, J.

M. Bulbous Plants: Biotechnology. P. 248-269.

Finnegan, E. J., Sheldon, C. C., Jardinaud, F., Peacock, W. J., & Dennis, E. S. (2004). A cluster

of Arabidopsis genes with a coordinate response to an environmental stimulus. Current

Biology, 14(10), 911-916.

Fortanier EJ (1973) Reviewing the length of the generation period and its shortening, particularly

in tulips. Scientia Horticulturae, 1(1):107-116

Kamenetsky, R., M. Zaccai, and M.A. Flaishman. (2012). Florogenesis, p. 197-232. In:

Kamenetsky, R. and H. Okubo (eds.). Ornamental geophytes: from basic science to

sustainable production. CRC press.

Leeggangers, H. A., Moreno-Pachon, N., Gude, H., & Immink, R. G. (2013). Transfer of

knowledge about flowering and vegetative propagation from model species to bulbous

plants. International Journal of Developmental Biology, 57(6-7-8), 611-620.

Michaels, S.D. and Amasino, R.M. (1999) FLOWERING LOCUS C encodes a novel MADS

domain protein that acts as a repressor of flowering. Plant Cell 11, 949–956

141

Miller, C.T., C.M. Filios, and W.B. Miller. (2012). Effects of flurprimidol, paclobutrazol and

uniconazole soaks and drenches on amaryllis (Hippeastrum) growth and development. Acta

Hort. 1002:431-438.

Srinivasan, C., Dardick, C., Callahan, A., & Scorza, R. (2012). Plum (Prunus domestica) trees

transformed with poplar FT1 result in altered architecture, dormancy requirement, and

continuous flowering. PLoS one, 7(7), e40715.

Roemer F (1907) Gladiolus praecox. Neue Gladiolen-Klasse, deren Sämlinge im ersten Jahre

blühen. Möller’s deutsche Gärtn. Ztg. 6:64-66

W. Atlee Burpee Company, Henry G. Gilbert Nursery and Seed Trade Catalog Collection (1913)

Burpee's annual 1914: the plain truth about seeds that grow. W. Atlee Burpee & Co.,

Philadelphia pp. 123

142

Appendices

A1. The primer design for VRN2 gene in the tested cereal and gladiolus samples.

Forward primer sequence Reverse primer sequence Annealing temperature (°C)

TCATCACCATCATCAGGA AAGCTTTTCTGGACTCGT 52

143

A2. Gel electrophoresis image of PCR VRN2 gene bands of 1500-2000 bp at 52oC annealing temperature. Rapid Generation Genotype (RGC) does not require vernalization to flower.

Gladiolus palustris is wild Eurasian species which require vernalization to flower, as well as all the cereal species require vernalization to flower. Both gladiolus samples do not show presence

VRN2 gene fragment amplification, although the primer design was based on the consensus sequence of VRN2 in Triticum monococum, Triticum durum, Hordeum vulgare and Aegilops tauschii. The failure to amplify VRN2 gene in gladiolus genotypes could indicate that gladiolus may have VRN2 gene homolog, but the gene sequence varies greatly to cereals.

144

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

A3. Gel electrophoresis image PCR VRN2 gene band of 1500-2000 bp at 51oC annealing temperature. Arranged from left wells to right; 1. High Ranger Plus 100 bp DNA ladder, 2.

Triticum monococum PI 272561 (control), 3. Gladiolus ×hybridus RGC genotype 2231, 4.

Gladiolus ×hybridus RGC genotype 1531151, 5. Gladiolus ×hybridus RGC genotype 16843, 6.

Gladiolus murielae, 7. Gladiolus cardinalis, 8. Gladiolus flanganii, 9. Gladiolus papilio, 10.

Freesia alba, 11. Iris sibirica, 12. Hippeastrum sp. 13. Lilium longiflorum ‘Nelly White’, 14.

Tulipa sp. ‘Queen of Night’, 15. Crocosmia ×crocosmiiflora ‘Mistral', 16. Crocus sativus, 17.

Cypella coelestis, 18. Allium cepa ‘Flat of Italy’, 19. Hyacinth orientalis ‘Purple voice’ 20.

Tecophilaea cyanocrocus. Thus, the designed primer was able to amplify VRN2 band in Triticum monococum while no band was amplified in any of geophytic genotypes, the designed primer was based on the consensus sequence of VRN2 in Triticum monococum, Triticum durum, Hordeum vulgare and Aegilops tauschii.

145

5 cm 5 cm

A4. The desirable height for Rapid Generation Cycling (RGC) genotypes for gladiolus potted production. RGC corm treated with high concentration of gibberellin inhibitor A-Rest® at 400 ppm (left). RGC seed planted and reached to flowering in less than a year (right).

146