Effects of day-length and temperature on floral structureand fertility restoration in a season-dependent male-sterileSolanum villosum Mill. mutant

Christopher O. Ojiewo Æ Kenji Murakami ÆPeter W. Masinde Æ Stephen G. Agong ÆMasaharu Masuda

Received: 6 November 2006 / Accepted: 17 April 2007 / Published online: 22 May 2007

� Springer Science+Business Media B.V. 2007

Abstract Solanum villosum is an important African

leafy vegetable whose yield is limited mainly by

competition from early and excess fruit-set. Induced

male-sterility is a potential tool to reduce this

competition and enhance yields. This study was

conducted to investigate the influence of photope-

riod and temperature on the floral dynamics of a

season-dependent male-sterile mutant. The mutant,

named T-5, has flowers which are sepaloid, mostly

stamenless, indeterminate and partially restored in

winter, late-spring, summer and autumn, respec-

tively. Floral organ restoration was found to be

largely independent of photoperiod conditions. Day/

night temperatures of 25/25 and 30/208C were found

to favour restoration of the floral organ but most

flowers were stamenless and infertile. High night

temperature favoured the formation of indeterminate

flowers both in the growth chamber (308C) and in

the greenhouse (>258C). On the other hand, low

growth chamber (108C) and greenhouse (<158C)

night temperature favoured the formation of sepal-

oid flowers. The optimum temperatures for floral

structure and fertility restoration were between 208Cand 258C (day) and 15–208C (night). Propagation of

T-5 mutant can thus be achieved by growing in

regions or seasons with such temperature ranges.

Under temperatures unfavourable for fruit-set, leaf

productivity is expected to be high.

Keywords Abnormal floral organ � Fertility

Restoration � Indeterminate flower

Introduction

Solanum nigrum complex is a conglomeration of

several Solanaceous taxa widely consumed as leafy

vegetables in most parts of Africa and South–East

Asia. The taxa in this section of African origin are

commonly referred to as African nightshade of which

S. villosum shows the widest distribution in the

tropics (Edmonds and Chweya 1997). The leaves

contain high levels of vitamins (especially A, B and

C), mineral fibres (such as iron, calcium and phos-

phorus), carbohydrates and proteins. They also con-

tain phenolics and alkaloids, such as nicotine,

quinine, cocaine, and morphine, which are known

for their medicinal attributes (Kokwaro 1993). The

demand for foods containing substances with phyto-

therapeutic (herbal medicine) activity is on the rise

C. O. Ojiewo � K. Murakami � P. W. Masinde �S. G. Agong � M. Masuda (&)

Graduate School of Natural Science and Technology,

Okayama University, 1-1-1 Tsushima Naka, Okayama,

Okayama 700-0080, Japan

e-mail: mmasuda@cc.okayama-u.ac.jp

P. W. Masinde � S. G. Agong

Department of Horticulture,

Jomo Kenyatta University of Agriculture and Technology,

P. O. Box 62000, Nairobi 00200, Kenya

123

Euphytica (2007) 158:231–240

DOI 10.1007/s10681-007-9445-z

worldwide (Duke 1990; Prohens et al. 2003). In

Kenya, as well as most parts of Africa and South–

East Asia, the perception of traditional leafy vegeta-

bles as ‘‘healthy’’ foods has accounted for a sudden

upsurge in their consumption, making them promis-

ing alternative cash crops locally and for export

(Schippers 2000).

Limited by a prolific early flowering and exces-

sive fruit-set which compete with leaf productivity,

the leaf yields of these important vegetables are,

however, relatively low. Induction and introduction

of male-sterile varieties has been proposed as an

economically viable and technically feasible strat-

egy to circumvent the post-anthesis sink-source

imbalance, improved and stabilize leaf yields

Ojiewo et al. (2005). We have induced four genic

male-sterile mutant types through seed irradiation

with c-ray namely: acetocarmine-stained unviable

pollen (T-1), defective pollen not stainable with

acetocarmine (T-2), pollenless (T-3), and low pollen

producing (T-4) types (Ojiewo et al. 2005). In

addition a novel male-sterile mutant with abnormal

floral organs (T-5) was isolated from one line

irradiated with 20 Gy 12C5+ (Ojiewo et al. 2006a).

Genic male-sterile lines pose a challenge in their

maintenance and propagation. Their maintenance

requires backcrossing with a heterozygote line, but

the progeny produced is 50% fertile and 50% male-

sterile. Partial fertility restoration of the M2 prog-

eny, desirable for maintenance and propagation,

was observed in some of the T-1*T-4 lines in

autumn, but the restoration was enhanced in the

subsequent M3 progeny, posing a threat in the use

of these lines as new varieties with high leaf yields

(Ojiewo et al. 2006b).

The abnormal floral organ mutant (T-5) has

exhibited a stable season-dependent structural

dynamics and fertility restoration. The M2, M3 and

M4 flowers were sepaloid from late-winter to mid-

spring, mostly stamenless in late-spring, indetermi-

nate in summer and partially restored with fruit-set

and seed-set in autumn. The objectives of the current

study were to (1) examine the possible influence of

day-length and/or temperature on this structural

dynamism and the eventual fertility restoration; (2)

determine the critical daylength and /or temperature

for fertility restoration to enable synchronizing of

seed (for propagation) and leaf production with

environmental conditions.

Materials and methods

Effects of daylength on floral organ structure and

fertility restoration

Seed from M3 generation of T-5 plants were sown in

cell trays containing vermiculite on 1 March 2005

and germinated seedlings raised in pots containing

commercial forest soil in a greenhouse. At floral bud

initiation, 3 plants each were transferred to 3 growth

chambers set at a constant day/night temperature of

25/258C and daylengths of 8, 12 and 16 h. All initial

floral buds were removed to give a uniform start. On

each of the plants, 15 floral trusses were sampled

from newly differentiating inflorescences and scored

for floral organ abnormalities, fruit-set and number of

seeds per fruit between 1 May and 15 June, 2005.

Although most flowers opened, fruit-set was negligi-

ble in all treatments. Many flowers at anthesis were

either stamenless or had distorted stamens, which

aborted a few days after anthesis.

In vitro pollen germination capacity, pollen

fertilization capacity on-stigma and stigma

viability

Pollen was collected at anthesis from normal looking

stamens and from greenhouse grown wild-type plants

for comparison. For in-vitro pollen germination, the

pollen grains were cultured in drops of a modified BK

medium (Brewbaker and Kwack 1963) consisting of

10% sucrose and 50 ppm H3BO3 in deionized

distilled H2O supplemented with 100 ppm CaCl2.

The pH of the medium was adjusted to 6.0. Hanging

drop cultures were prepared by placing three 30 ll

drops of the growth medium on 30 mm Petri-dishes.

The anthers were held with forceps and gently tapped

onto each drop to release small amounts of pollen.

The Petri dishes were then gently overturned onto

Petri dish covers with moist filter papers. Pollen was

left to germinate in an incubator set at 258C and

observed under a light microscope after 1.5*2 h.

Pollen was considered to have germinated when the

pollen tube was equal to or more than the pollen grain

diameter (Heslop-Harrison et al. 1984). For each

drop, about 100 pollen grains were scored and for

each daylength treatment three replicates were used.

Flowers from 5 trusses each of 3 wild-type plants

raised in the greenhouse were emasculated 2 days

232 Euphytica (2007) 158:231–240

123

before anthesis and pollinated at anthesis with pollen

from each of the daylength treatments. At the same

time, 5 floral trusses were selected from each

daylength treatment and stamenless flowers polli-

nated with pollen from the wild-type plants. The rate

of fruit-set, number of seeds per fruit and the

germination capacity of F1 seed resulting from the

each pollination were examined.

Effect of temperature on floral organ structure and

fertility restoration

To examine the effect of night temperature, plants

maintained by cutback were transferred to growth

chambers where night temperatures were set at 30, 20

and 108C. A neutral daylength (12/12 h) was

maintained and day temperature was set at 308C in

all the 3 chambers. There was no fruit-set in any of

the set-ups. To determine the optimum day/night

temperature integrals for fruit-set, new seedlings

were raised as above in a greenhouse where mini-

mum temperature was maintained at 128C between

15 Jan and 28 Feb 2006. Just before flower bud

initiation, three plants each were transferred to 3

separate growth chambers where a neutral daylength

(12/12 h) was maintained and day/night temperatures

were set at 25/20, 25/15 and 20/158C, respectively.

The rate of occurrence of floral organ abnormalities

and fruit- and seed-set in restored flowers were scored

from inflorescences sampled as above. The remaining

plants were raised in the greenhouse where similar

data was collected from 3 plants at the beginning,

middle and end of March, April and May. In addition,

in vitro pollen germination capacity of greenhouse

raised T-5 plants was investigated and compared with

wild-type normal pollen.

Statistical analysis

With only 3 phytochambers each with a different

lighting and temperature regime there was no true

replication in this experiment, so variability between

plants within compartments was used as a proxy

error. The experiment assumes a completely random-

ized design (CRD) with 3 replicate plants per

treatment. For ease and consistency in data analysis,

a similar set up was repeated in the greenhouse

experiments. All statistical analyses were performed

using SPSS. Analysis of variance (ANOVA) was

used to determine if the means of the treatments were

significantly different. Discrete variables, such as the

number of seeds per fruit, were analyzed after square

root transformation and proportions were analyzed

after logarithmic transformation to stabilize the

variance among groups. Differences in the parame-

ters measured between the different lighting and

temperature regimes were evaluated based on LSD

(P � 0.05). Results are presented as the mean ± SE

for n = 3. Means within the same column and

experiment followed with different letters were

significantly different.

Results

Effects of daylength on floral organ structure and

fertility restoration

All floral buds in the growth chamber as well as the

greenhouse were sepaloid at initiation regardless of

daylength or temperature treatment. While the wild-

type flowers opened 2–3 days after flower bud

initiation, T-5 floral organ differentiation occurred

2 weeks after initiation of the sepaloid buds. Some

flower buds aborted before any differentiation was

observed, while others aborted at ‘anthesis’ (Fig. 1).

However, all computations were based on the orig-

inally initiated flower buds (5–8 flowers per inflores-

cence) and the differences discussed on the basis of

their significance in agricultural application.

While maintaining the day and night temperature

constant at 258C and only varying the photoperiod,

Fig. 1 Wild-type floral

buds at anthesis (a), T-5

flowers initiated at the same

time as the wild-type still in

sepaloid state (b) and

differentiating T-5 flowers

after 2 weeks (c)

Euphytica (2007) 158:231–240 233

123

there were negligible incidences of sepaloid or

indeterminate flowers (Table 1). Apart from a few

flowers that aborted, more than 90% of the remaining

flowers opened, regardless of the photoperiod. How-

ever, considering that fruit-set of wild-type plants in

the greenhouse was perfect (100%), the fruit-set in

the open T-5 flowers was considered negligible

(0*4%) in all the 3 set-ups. Most of the fruits were

malformed with no seeds even at maturity (Fig. 2a, b)

while the seeded fruits (Fig. 2c, d) had very few seeds

per fruit (9*13) compared to >60 seeds per fruit in

wild-type plants.

A closer observation of the floral structure

revealed that 30–40% of the flowers had malformed

stamens (Fig. 3a) that dropped off within 3 days after

anthesis. A significant proportion of the opening

flowers were, thus, left stamenless. In addition, the

unaborted stamens exhibited other abnormalities such

as fusion to petals (Fig. 3b) or fusion to carpels

(Fig. 3c). Besides, there were also malformations

such as splitting of the styles and stigmas (Fig. 3d),

unfused multiple styles with no distinct stigmas

(Fig. 3e) and unfused multiple styles and stigmas

(Fig. 3f). To this end it was hypothesized that failure

of fruit-set was due to these stamen and carpel

abnormalities.

In vitro pollen germination capacity, pollen

fertilization capacity on-stigma and stigma

viability

To test the viability of the T-5 pollen, in-vitro

germination tests were carried out and compared

with wild-type pollen. At the same time the pollen

was used to pollinate emasculated flowers of wild-

type plants in the greenhouse. To test the viability

of the T-5 carpels, stamenless flowers were polli-

nated with wild-type pollen. T-5 pollen germination

Table 1 Effect of daylength on the occurrence of floral organ abnormalities, fruit- and seed-set

Daylength (hr)z Frequency of abnormalities (%) Fruit-set (%) No. seeds per fruit

Sepaloid Stamenless Indeterminate

8 1.3 ± 0.0a 40.2 ± 2.3a 7.0 ± 0.2a 0.0 –

12 3.1 ± 0.1a 30.5 ± 1.8b 4.2 ± 0.1a 1.4 ± 1.0a 13.2 ± 0.6a

16 3.5 ± 0.1a 40.8 ± 2.1a 0.0 3.9 ± 1.2a 9.3 ± 1.1a

z Growth chamber day and night temperature was maintained constant at 258C throughout the experiment. Data are means ± SE;

n = 3. Means within the same column sharing the same letters are not significantly different based on LSD0.05

Fig. 2 Fruit abnormalities

observed under constant

day/night temperature of

25/258C with varying

photoperiods. Some

malformed fruits with

vegetative parts attached (a)

remained small and seedless

even at maturity (b), while

others had very few seeds

(c and d)

234 Euphytica (2007) 158:231–240

123

in-vitro was significantly lower (11*22%) than that

of wild-type pollen (55*61%). In addition, fruit-set

with T-5 pollen on W-T flowers was considerably

low (16*27%), with only a few seeds per fruit

(12*19). On the other hand, there was a high rate

of fruit-set (85*94%) when wild-type pollen was

used to pollinate the open T-5 flowers, with the

mean number of seeds per fruit ranging from 48 to

63. There were minimal fruit abnormalities (Fig. 4b,

c), though not ultimately absent. Save for many

large sepals still attached to the fruit-stalk, most of

the fruits were properly formed (Fig. 4d) with a

near wild-type fruit shape (Fig. 4a). The germina-

tion capacity of both seed lots was high (84*91%)

with no differences of agricultural significance

(Table 2).

Effect of temperature on floral organ structure and

fertility restoration

There were significant differences (P < 0.05) between

the three night temperature regimes with regard to

floral organ structure abnormalities and its restora-

tion. Maintaining the daylength at 12 h and the day

temperature at 308C, high night temperature (308C)

resulted mostly in indeterminate flowers, low night

temperature (108C) in sepaloid flowers, while mod-

erate night temperature (208C) resulted in some

anthesis (Table 3). Open flowers under 30/208C had

similar abnormalities observed during the photope-

riod experiment under 25/258C, with 13.3% of the

flowers being stamenless. However, there was no

fruit-set in any of the 3 night temperature set-ups.

With day and night temperature integrals of 25/158Cand 20/158C and 25/208C, there was minimal flower

abortion before anthesis. Although all initial flowers

were sepaloid, floral organ differentiation occurred

within 2 weeks as expected with no flowers retaining

the sepaloid form, or becoming indeterminate and the

frequency of stamenless open flowers was negligible

(1–2%). Although not comparable to the wild-type

plants, fruit-set in all the three set-ups was signifi-

cantly high (40*50%), with 25/208C being the

highest. At the same time, cases of fruit malformation

were minimal. The mean number of seeds per fruit

was also significantly higher (&40) than the earlier

set-ups (Table 4). There were no significant differ-

ences in the floral organ restoration and the rate of

fruit- and seed-set among the 3 day and night

temperature integrals.

Fig. 3 Types of stamen and

carpel abnormalities

observed under day/night

temperature of 30/208C and

25/258C with various

photoperiods. Top row, left

to right: stamen abortion

(a), fusion to petals (b) and

fusion to carpels (c). Lower

row, left to right: carpel

splitting (d), multiple

carpels with no distinct

stigmas (e) and unfused

carpels and stigmas (f)

Fig. 4 Varying degrees of fruit abnormalities (b and c) observed when normal pollen was used to pollinate T-5 plants raised under

25/258C with various photoperiods. Most fruits were normal (d) and comparable to wild type fruits (a)

Euphytica (2007) 158:231–240 235

123

In the greenhouse, the frequency of sepaloid

flowers was high at the beginning of March and

continued to reduce up to late April when the night

temperature was still <158C (Table 5). However,

from early May when the night temperature had risen

beyond 158C, no flower retained the sepaloid form

beyond 2 weeks after initiation of floral buds.

Stamenless flowers were observed from mid-April

and the frequency continued to increase up to early-

May, and then reduced thereafter. Indeterminate

flowers started to appear in early-May when the day

temperatures had risen above 258C and continued to

increase thereafter as summer approached. In-vitro

pollen germination capacity and the corresponding

fruit-set increased from early-March to late-April,

and then decreased thereafter. The number of seeds

per fruit followed the same trend but started declining

between early and mid April and continued in the

same trend till the end of the experiment in late May.

Incidentally, the proportion of malformed fruits

increased concomitantly with increase in the green-

house temperature from early March to late May,

with the latest fruits being seedless (Fig. 5a), fol-

lowed by no fruit-set (Fig. 5b).

Discussion

Many species have an obligatory requirement for the

correct photoperiod for transition from vegetative to

reproductive phase. Flowering in the plant model

Arabidopsis occurs much earlier under long days of

Table 2 Effect of daylength on in-vitro pollen germination, fertilization and seed germination capacity

Source of T-5 Cross In vitro pollen germination (%)z Fruit-set (%) No. of seeds per fruit Seed germination (%)

8h T-5 x WT 61.2 ± 2.6a 85.4 ± 4.2a 56.3 ± 3.3ab 84.3 ± 2.9a

WT x T-5 11.3 ± 1.1c 16.8 ± 1.2c 13.2 ± 1.4c 90.7 ± 2.4a

12h T-5 x WT 58.7 ± 1.6a 94.1 ± 2.6a 48.9 ± 4.1b 89.3 ± 5.6a

WT x T-5 13.1 ± .08c 27.5 ± 3.4b 19.4 ± 2.4c 89.7 ± 2.3a

16h T-5 x WT 55.5 ± 2.2a 89.5 ± 6.7a 62.8 ± 5.6a 89.3 ± 2.8a

WT x T-5 22.4 ± 2.3b 24.3 ± 1.7bc 12.6 ± 2.3c 87.0 ± 6.4a

z This is the pollen germination of the pollen parent (not the seed parent). Data are means ± SE; n = 3. Means within the same

column sharing the same letters are not significantly different based on LSD0.05

Table 3 Effect of night temperature on the occurrence of floral organ abnormalities, fruit- and seed-set

Night temp. (8C)z Frequency of abnormalities (%) Fruit-set No. seeds per fruit

Sepaloid Stamenless Indeterminate

10 75.8 ± 9.7a 0.0 14.2 ± 1.9a 0.0 –

20 34.6 ± 3.6b 13.3 ± 3.6 28.4 ± 2.1b 0.0 –

30 24.3 ± 5.6c 0.0 65.7 ± 3.4c 0.0 –

z Growth chamber day temperature was maintained constant at 308C while the daylength was neutral (12 h) throughout the

experiment. Data are means ± SE; n = 3. Means within the same column followed by different letters are significantly different based

on LSD0.05

Table 4 Combined effect of day and night temperature on the occurrence of floral organ abnormalities, fruit- and seed-set

Temperature (oC) Frequency of abnormalities (%) Fruit-set (%) No. seeds per fruit

Day (12h) Night (12h) Sepaloid Stamenless Indeterminate

20 15 0.0 2.1 ± 0.3a 0.0 38.6 ± 4.9a 37.0 ± 2.5a

25 15 0.0 1.4 ± 0.1a 0.0 41.7 ± 5.6a 38.7 ± 3.6a

25 20 0.0 0.0 0.0 54.5 ± 3.3b 37.5 ± 4.0a

236 Euphytica (2007) 158:231–240

123

16 h light than under short days of 8 h, and

daylengths between these two extremes give an

intermediate response (Reeves and Coupland 2000).

Mutations causing insensitivity to photoperiod on

flowering have been described, and one single genetic

pathway has been identified in an analysis of allelic

differences between ecotypes (Koornneef et al.

1998). On the other hand, mutations causing sensi-

tivity to changes in photoperiod conditions in

otherwise day-neutral plants such as tomato have

also been reported (Sawhney 2004). While the wild-

type plants of the taxa belong to the S. nigrum

complex are not known to respond to seasonal

changes in temperature or photoperiod, the flowers

of T-5 mutant used in this study vary in structure and

fertility with the four seasons of the year. Near

perfect restoration of the floral organ structure in T-5

was observed in similar degrees under all daylength

conditions in the growth chamber. However, a

combination of abnormalities in the floral structure

such as stamen abortion or homeotic transformation

to adjacent organs, as well as low pollen viability of

restored flowers resulted in very low rate of fruit-set

under all daylength conditions. There were slight

differences of no agricultural significance in the rate

of fruit-set and number of seeds per fruit between the

daylength treatments. In addition, most of the few

fruits were malformed and sometimes seedless. The

stigma was, however, found to be viable and

compatible with wild-type pollen at all daylengths.

Since these defects were observed in all the 3

daylengths with agriculturally insignificant differ-

ences, it was concluded that photoperiod had only a

slight effect on the flowering habit of T-5 mutant.

Table 5 Effect of greenhouse air temperature on the occurrence of floral organ abnormalities, in-vitro pollen germination, fruit- and

seed-set

Sampling

periodyMean temp

(oC)zFrequency of abnormalities

(%)

In-vitro pollen

germination (%)

Fruit-set (%) No. of seeds

per fruit

Day Night Sepaloid Stamenless Indeterminate

March Early 17.2 8.4 87.3 ± 2.5a 0.0 0.0 14.9 ± 4.8a 11.4 ± 2.1a 10.8 ± 4.5a

Mid 16.8 7.8 85.5 ± 2.6a 0.0 0.0 18.5 ± 3.9a 12.0 ± 4.7a 13.0 ± 3.3a

Late 19.4 9.3 68.1 ± 2.9b 0.0 0.0 20.7 ± 2.3ab 23.4 ± 2.2b 18.4 ± 3.9ab

April Early 21.4 12.8 63.9 ± 3.4b 0.0 0.0 28.6 ± 4.8bc 33.2 ± 3.4cd 32.8 ± 3.6c

Mid 22.6 14.4 36.5 ± 5.3c 20.5 ± 2.7a 0.0 35.2 ± 1.9cd 39.8 ± 3.9d 32.6 ± 2.9c

Late 24.1 14.2 28.3 ± 3.6c 23.8 ± 3.8a 0.0 42.4 ± 4.2d 38.8 ± 4.5d 21.0 ± 1.1b

May Early 29.7 19.7 0.0 23.0 ± 2.6a 17.6 ± 2.1a 23.3 ± 2.2ab 23.6 ± 4.7b 17.8 ± 2.2ab

Mid 27.6 19.3 0.0 25.4 ± 2.3a 38.5 ± 3.2b 18.9 ± 3.4a 19.0 ± 3.6ab 13.8 ± 1.2a

Late 32.1 21.1 0.0 9.9 ± 0.7b 67.2 ± 4.2c 15.7 ± 2.0a 18.8 ± 2.1ab 10.8 ± 2.8a

y Analysis of data was done independent of the sampling period (i.e. only temperature effect considered). Data are means ± SE;

n = 3. Means within the same column followed by different letters are significantly different based on LSD0.05

z This is mean temperature of about 10 days of the early, mid and late parts of each month

Fig. 5 Extremities of fruit

abnormality observed as

temperature rises towards

end of spring (a) and the

subsequent indeterminate

flowers without fruit-set in

early summer (b)

Euphytica (2007) 158:231–240 237

123

Further, it was postulated that the seasonal changes in

T-5 flower were dependent on temperature, and that a

constant day/night temperature of 25/258C could

favour partial floral organ structure but not fertility

restoration.

Temperature conditions are known to influence

the development of floral organs and sex expression

in many species of flowering plants. When tomato

and pepper plants (two Solanaceous vegetables)

were exposed to low temperatures, they produced

flowers showing alterations in the number, morphol-

ogy, and pattern of fusion of floral organs. Conse-

quently, malformed fruits were produced from these

flowers (Sawhney 1983; Polowick and Sawhney

1985). However, there were striking differences in

the response of tomato and pepper to low temper-

ature. For example, in pepper flowers, low temper-

ature inhibited the enlargement of petals and

stamens but enhanced the same in tomato. In

addition, low temperature induced male-sterility in

pepper but not in tomato. On the other hand, low

temperature resulted in increased number of petals,

stamens and locules in tomato but not in pepper.

These results suggest that temperature conditions

may induce different responses in different species

within the same family. Our results under day/night

temperatures of 258C and 30/208C have some

similarities to these observations, though the tem-

perature was not low. However, at low temperatures

our results are completely different from these

reports.

Work with Arabidopsis thaliana and Antirrhinum

majus has allowed the genetic and molecular char-

acterization of several families of regulatory genes

that control flower initiation and development. In

these species and many flowering plants, floral organ

identity is regulated by homeotic functions of genes

mostly of the MADS-box class. A gene-based

‘‘ABCDE’’ model outlining the overlapping activi-

ties of five classes of regulatory genes, which are

responsible for the identity of the five floral organs

has been developed: A and E for the sepals; A, B and

E for the petals; B, C and E for the stamens; C and E

for the carpels; and C, D and E for the ovules

(Theissen 2001). Mutations affecting the proper

function of any of these genes promote homeotic

transformations and, consequently, abnormal floral

organogenesis. Temperature has been reported to

affect the function of floral structure regulatory genes

in Antirrhinum (Zachgo et al. 1995) and Arabidopsis

(Sablowski and Meyerowitz 1998). In both cases,

mutant plants grown at low temperature (15–168C)

produce nearly wild-type flowers, but growth at the

nonpermissive temperature (26*288C) causes the

second-whorl organs to develop as sepals (only A

function present) and stamens to be replaced by

carpelloid organs (only C function left in the third

whorl). In this study, both 30/208C and 25/258C were

found to favour anthesis in T-5 but not fruit-set. High

(308C) and low (108C) night temperature extremes

hampered flower restoration, with the former favour-

ing indeterminate flower development while the latter

favoured sepaloid flowers. An interplay between

night and day temperature was necessary to achieve

both floral organ and fertility restoration, the opti-

mum temperatures being between 208C–258C (day)

and 15–208C (night). The response of the floral organ

development and fertility restoration was similar in

the greenhouse as in the growth chamber.

We have previously described the homeotic

changes in T-5 mutant flower in comparison to the

wild-type flowers (Ojiewo et al. 2006a). As expected

for homeotic mutants, these changes reproduced the

organ identity of adjacent whorls and the intermedi-

ate organs formed kept the growth pattern character-

istic of the whorl they occupy. The observed

homeotic transformations followed normal patterns

in their time of appearance, the kind of transforma-

tion, and the whorls affected. Regarding the time of

appearance, the homeotic transformations took place

late in T-5 floral organ development, giving rise to

chimeric organs such as petaloid sepals and capelloid

stamens. As a result there were more carpels than the

standard single carpel in wild-type flower. However,

observations from this study suggest that splitting

events taking place in the carpel could also contribute

to this phenomenon. Meristematic mutations altering

the number of floral organs have been reported in

Arabidopsis. Genes such as CLAVATA1 (CLV1) and

CLAVATA3 (CLV3) seem to be required to regulate

the size of the shoot and floral meristems, which in

turn affects the number of floral organs that develop

in each whorl. Weak alleles at loci clv1 and clv3

plants develop enlarged shoot and floral meristems,

which give rise to flowers with additional organs in

each whorl (Clark et al. 1993, 1995, 1997). These

phenotypic similarities could suggest an influence of

temperature on the expression or activity of the

238 Euphytica (2007) 158:231–240

123

corresponding T-5 genes in flowers restored under

30/25 or 25/258C. Unfortunately, T-5 orthologs of

any floral meristem or floral organ identity genes

have not been isolated, which precludes any analysis

of the effects of temperature on their expression at the

moment.

Zachgo et al. (1995) reported that sensitivity to

temperature in Antirrhinum def-101 was a property of

a mutant protein which loses its active conformation

or the ability to interact functionally with other

proteins under non-permissive temperatures. On the

other hand, Sablowski and Meyerowitz (1998) dem-

onstrated that temperature sensitivity in Arabidopsis

ap3–1 mutant was caused by a temperature-sensitive

splicing defect resulting in unstable RNA–RNA

interactions. Since there are no reports (to the best

of our knowledge ) of any mutant plant species whose

floral organ structure varies from season to season

with accurate repetition of the pattern every year, the

dynamic nature of the T-5 flower structure is of

potential biochemical and physiological significance

in floral genetics. On the other hand, these charac-

teristics are of various potential applications from a

breeding and agronomic point of view. Seed produc-

tion for maintenance and propagation purposes can be

done by temporal or spatial manipulation to synchro-

nize flowering with temperatures of 20–258C (day)

and 15–208C (night). Taking the case of Kenya (and

most parts of Africa) where the wild-type S. villosum

was obtained, this can be achieved by producing T-5

plants in cooler months of the year such as June–

August or in highland regions which remain cool

throughout the year. When and where fruit-set is

hampered (sepaloid, stamenless, and indeterminate),

the competition between the vegetative and repro-

ductive function is eliminated hence leaf yield is

expected to be high. Regardless of photoperiod

conditions, hybrid seed production using T-5 as the

female parent could potentially be done under

temperatures that favour stamenless flower formation

with minimum contamination risk. Environmental

modulation of male-sterility has been useful in plant

breeding (reviewed by Perez-Prat and Campagne

2002). For example, the use of thermo-photosensitive

genic male-sterile mutants provides the opportunity

to produce rice hybrid seed by a two-line system. For

maintenance, the male-sterile female line can be

propagated by growing it under environmental con-

ditions that restore fertility (Wu et al. 2003).

Conclusions and recommendations

As there were no significant differences between 8,

12 and 16 h daylength with regard to floral organ

structure restoration, fruit- and seed-set, it is plausible

to conclude that the African nightshade T-5 mutant is

photoperiod-insensitive. The results suggest rather

that it is temperature-sensitive, the critical minimum

and maximum temperatures for floral organ and

fertility restoration being 158C and 258C, respec-

tively. The male function of the T-5 mutant flower

can be eliminated when not needed by growing in

restrictive temperatures and restored when needed by

growing in permissive temperatures. Thus, produc-

tion of fruit and seed (which are the major sink of the

plant) can be eliminated to improve leaf productivity

and restored for purposes of propagation. Further

analysis of wild-type and T-5 mutant through gene

cloning and sequencing, gene expression studies and

phylogenetic constructions is necessary to identify

the mutant lesions resulting in the T-5 phenotype. In

addition, the causes of temperature sensitivity in the

affected gene and its interaction with other genes

controlling floral organ development and final iden-

tity should be investigated further.

References

Brewbaker JL, Kwack BH (1963) The essential role of calcium

ion in pollen germination and pollen tube growth. Am J

Bot 50:859–886

Clark SE, Running MP, Meyerowitz EM (1993) CLAVATA1, a

regulator of meristem and flower development in Ara-bidopsis. Development 119:397–418

Clark SE, Running MP, Meyerowitz EM (1995) CLAVATA3 is

a specific regulator of shoot and floral meristem devel-

opment affecting the same processes as CLAVATA1.

Development 121:2057–2067

Clark SE, Williams RW, Meyerowitz EM (1997) The CLAV-ATA1 gene encodes a putative receptor kinase that con-

trols shoot and floral meristem size in Arabidopsis. Cell

89:575–585

Duke JA (1990) Promising phytomedicinals. In: Janick J, Si-

mon JE (eds) Advances in new crops. Portland, Timber

Press, OR, USA, pp. 491–498

Edmonds JM , Chewya JA (1997) Black night shades Solanumnigrum L. and related species promoting the conservation

and the use of underutilized and neglected crops 15.

Institute of Plant Genetics and Crop Plant Research,

Gatersleben/international Plant Genetic Resources Insti-

tute, Rome, Italy

Heslop-Harrison J, Heslop-Harrison Y, Shivanna KR (1984)

The evaluation of pollen quality and a further appraisal of

Euphytica (2007) 158:231–240 239

123

the fluorochromatic (FCR) test procedure. Theor Appl

Genet 67:367–375

Kokwaro JO (1993) Medicinal plants of East Africa, 2nd edn.

Kenya Literature Bureau, Nairobi, pp. 58, 157, 224

Koornneef M, Alonso-Blanco C, Blankestijn-de Vries H,

Hanhart CJ, Peeters AJM (1998) Genetic interactions

among late flowering mutants of Arabidopsis. Genetics

148:885–892

Ojiewo CO, Agong SG, Murakami K, Tanaka A, Hase Y,

Masuda M (2005) Male-sterility induced by c-ray irradi-

ation of African nightshade (Solanum nigrum L. ssp.

villosum) seed. J Horticult Sci Biotechnol 80:699–704

Ojiewo CO, Agong SG, Murakami K, Tanaka A, Hase Y

Masuda M (2006a) Biological effect of carbon-ion beam

on mutagenesis and a novel season-dependent floral

homeotic mutant in Solanum villosum Miller. J Horticult

Sci Biotechnol 81:559–564

Ojiewo CO, Murakami K, Masuda M (2006b) A novel male-

sterile mutant with season-dependent floral homeotic trait

induced by 12C5+ ion beam irradiation of African night-

shade (Solanum villosum Mill) seed. Presented at the 27th

International Horticultural Congress. COEX Convention

Centre, Seoul, Korea, 13th–19th August, 2006. IHC 2006

Abstracts. S01-P-139 p 43

Perez-Prat E, Campagne MML (2002) Hybrid seed production

and the challenge of propagating male-sterile plants.

Trends Plant Sci 7:199–203

Polowick PL, Sawhney VK (1985) Temperature effects on

male-fertility and flower and fruit development in Cap-sicum annuum L. Scientia Horticulturae 25:117–127

Prohens J, Rodrı́guez-Burruezo A, Nuez F (2003) New crops:

an alternative for the development of horticulture. Food

Agric Environ 1:75–79

Reeves PH, Coupland G (2000) Response of plant development

to environment: control of flowering by daylength and

temperature. Current Opin Plant Biol 3:37–42

Sawhney VK (1983) Temperature control of male sterility in a

tomato mutant. J Heredity 74:51–54

Sablowski RWM, Meyerowitz EM (1998) Temperature-sensi-

tive splicing in the floral homeotic mutant apetala3–1.

The Plant Cell 10:1453–1463

Sawhney VK (2004) Photoperiod-sensitive male-sterile mutant

in tomato and its potential use in hybrid seed production. J

Horticult Sci Biotechnol 79:138–141

Schippers RR (2000) African indigenous vegetables. An

overview of the cultivated species. Revised CD ROM

Edition. Horticultural Development Services, Chatham,

UK, pp. 25–31, 147–192

Theissen G (2001) Development of floral organ identity: stories

from the MADS house. Current Opin Plant Biol 4:75–85

Wu D, Shen S, Cui H, Xia Y, Shu Q (2003) A novel thermo/

photoperiod-sensitive genic male-sterile (T/PGMS) rice

mutant with green-revertible albino leaf colour marker

induced by gamma irradiation. Field Crops Res 81:141–

147

Zachgo S, Silva EA, Motte P, Trobner W, Saedler H, Schwarz-

Sommer Z (1995) Functional analysis of the Antirrhinumfloral homeotic DEFICIENS gene in vivo and in-vitro

using a temperature sensitive mutant. Development

121:2861–2875

240 Euphytica (2007) 158:231–240

123