3pages for agriculture paper due by 24 hours
Golden bananas in the field: elevated fruit pro-vitamin A from the expression of a single banana transgene Jean-Yves Paul1, Harjeet Khanna1,†, Jennifer Kleidon1, Phuong Hoang1, Jason Geijskes1,‡, Jeff Daniells2, Ella Zaplin1,§, Yvonne Rosenberg3, Anthony James1, Bulukani Mlalazi1, Pradeep Deo1, Geofrey Arinaitwe4, Priver Namanya1,4, Douglas Becker1, James Tindamanyire1, Wilberforce Tushemereirwe4, Robert Harding1 and James Dale1,*

1Centre for Tropical Crops and Biocommodities, Queensland University of Technology, Brisbane, Qld, Australia 2Agri-Science Queensland, Department of Agriculture and Fisheries, South Johnstone, Qld, Australia 3PlantVax Inc, Rockville, MD, USA 4National Agricultural Research Laboratories, National Agricultural Research Organization, Kampala, Uganda

Received 15 August 2016;

revised 6 October 2016;

accepted 7 October 2016.

*Correspondence (Tel +61 7 3138 2819; fax

+61 7 3138 4132; email [email protected])

Present addresses: †Sugar Research Australia,

Brisbane, Qld, Australia. ‡Syngenta Asia Pacific, Singapore,

Singapore. §Charles Sturt University, Wagga Wagga,

NSW, Australia.

Keywords: Vitamin A deficiency,

Uganda, pro-vitamin A, staple food

crop, banana, biofortification, genetic

modification.

Summary Vitamin A deficiency remains one of the world’s major public health problems despite food

fortification and supplements strategies. Biofortification of staple crops with enhanced levels of

pro-vitamin A (PVA) offers a sustainable alternative strategy to both food fortification and

supplementation. As a proof of concept, PVA-biofortified transgenic Cavendish bananas were

generated and field trialed in Australia with the aim of achieving a target level of 20 lg/g of dry weight (dw) b-carotene equivalent (b-CE) in the fruit. Expression of a Fe’i banana-derived phytoene synthase 2a (MtPsy2a) gene resulted in the generation of lines with PVA levels

exceeding the target level with one line reaching 55 lg/g dw b-CE. Expression of the maize phytoene synthase 1 (ZmPsy1) gene, used to develop ‘Golden Rice 2’, also resulted in increased

fruit PVA levels although many lines displayed undesirable phenotypes. Constitutive expression

of either transgene with the maize polyubiquitin promoter increased PVA accumulation from the

earliest stage of fruit development. In contrast, PVA accumulation was restricted to the late

stages of fruit development when either the banana 1-aminocyclopropane-1-carboxylate oxidase

or the expansin 1 promoters were used to drive the same transgenes. Wild-type plants with the

longest fruit development time had also the highest fruit PVA concentrations. The results from

this study suggest that early activation of the rate-limiting enzyme in the carotenoid biosynthetic

pathway and extended fruit maturation time are essential factors to achieve optimal PVA

concentrations in banana fruit.

Introduction

Micronutrient deficiency, often referred to as hidden hunger,

occurs when intake and absorption of vitamins and minerals are

too low to sustain good health and development. The World

Health Organization (WHO) estimates that 190 million pre-school

children are deficient in one of the major micronutrients, vitamin

A. Vitamin A deficiency (VAD) alone is responsible for almost 6%

of child deaths under the age of 60 months in Africa and 8% in

South-East Asia (WHO, 2011). For the vast majority of these

children, VAD is almost exclusively the result of inadequate intake

of dietary vitamin A or pro-vitamin A (PVA) although exacerbated

by other health conditions. Similar levels of VAD are also evident

in women of childbearing age in these same regions (WHO,

2011). These levels of VAD continue despite the implementation

over many years of extensive alleviating strategies such as

supplements and food fortification. These strategies have been

demonstrably successful, but there remain persistently high and

unacceptable levels of VAD particularly in sub-Saharan Africa and

south Asia (Stevens et al., 2015).

In an effort to significantly reduce VAD in these regions,

strategies aimed at increasing the dietary intake of particularly a- and b-carotene together as PVA are being developed or imple- mented. These include programmes to encourage growing and

consuming staple foods with high levels of PVA. In some

instances, such foods or crops with the desired agronomic and

consumer traits are already available and can therefore be easily

deployed (HarvestPlus, 2012). However, the majority of accepted

cultivars and landraces of staple crops are low in micronutrients

such as PVA and iron, and therefore, it is necessary to develop

new varieties with enhanced levels of these micronutrients. This

can be achieved either through conventional breeding or by

genetic modification where the traits are not available within the

accessible germplasm or cannot be easily introgressed into

acceptable cultivars. These two approaches are known as

biofortification.

The best-known example of biofortification by genetic mod-

ification, and the most advanced in terms of development, is

‘Golden Rice’. In Ye et al., 2000 and colleagues reported the

generation of transgenic rice expressing the daffodil phytoene

synthase (Psy) gene under the control of an endosperm-specific

rice glutelin promoter together with the bacterial (Pantoea

ananatis formerly known as Erwinia uredovora) phytoene desat-

urase (CrtI) gene under the control of the constitutive CaMV 35S

promoter. The endosperm of selected lines was yellow, and one

heterozygous line contained 1.6 lg/g dry weight (dw) total carotenoids. Paine et al. (2005) subsequently reported the

development of the second generation of ‘Golden Rice’, which

520 ª 2016 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution License, which permits use,

distribution and reproduction in any medium, provided the original work is properly cited.

Plant Biotechnology Journal (2017) 15, pp. 520–532 doi: 10.1111/pbi.12650

http://creativecommons.org/licenses/by/4.0/
was engineered with a maize (Zea mays) Psy gene and the CrtI

gene under the control of the glutelin promoter. One Golden Rice

2 elite event had a 23-fold increase in total carotenoids over the

original Golden Rice with a total carotenoid level of up to 37 lg/g dw in the endosperm of which 31 lg/g was b-carotene. A number of other important food crops have been or are being

developed to enhance the level of PVA through genetic modi-

fication.

Bananas are the world’s most important fruit crop and one of

the top 10 crops by production. They are widely grown in the wet

tropics and subtropics forming an important dietary component

both raw as a dessert fruit and cooked often as the major source

of carbohydrate. In a number of countries, bananas are the

principal staple food including Uganda where consumption levels

average 0.5 kg per person per day rising to around 1 kg per

person per day in some regions (Komarek, 2010; Smale and

Tushemereirwe, 2007). In East Africa, the staple cultivar is East

African highland banana (EAHB) (M. acuminata AAA-EA) pre-

pared primarily by steaming or boiling whereas, in West Africa,

plantains are dominant and are usually fried or roasted (Fungo

and Pillay, 2011). In both regions, the level of VAD is high. In

Uganda, it varies from 15% to 33% in children under 60 months

with similar levels in women of childbearing age (UDHS – Uganda

Demographic and Health Survey, 2006). Unlike rice endosperm,

banana fruit contains PVA and, in some instances, very high levels

particularly Fe’i bananas of Micronesia and Papua New Guinea

(Englberger et al., 2003). Bananas with b-carotene equivalent (b- CE) levels of 340 lg/g dw have been reported whereas the dominant dessert banana cv ‘Cavendish’ has between 1 and

4 lg/g dw b-CE and the EAHB clone, ‘Nakitembe’, has approx- imately 10 lg/g b-CE dw (Englberger et al., 2006a; Mbabazi, 2015). Unfortunately, domesticated bananas have very low male

and female fertility rendering conventional breeding extremely

difficult. Thus, the introgression of the high PVA traits of Fe’i

bananas for instance into farmer preferred EAHB selections would

be practically impossible. However, genetic modification of

bananas is well established.

Here, we report the ‘proof-of-concept’ technology required

towards the generation of PVA-biofortified EAHB varieties in

Uganda. The ‘Cavendish’ dessert banana was genetically modi-

fied, and greatly enhanced PVA levels were demonstrated in the

fruit of plants grown in the field in Australia.

Results

The target

At the outset, it was important to identify a target fruit level of b- carotene equivalents necessary to help alleviate VAD in Uganda.

The target was set at delivering 50% of the estimated average

requirement (EAR) of vitamin A in vulnerable populations which

for children under 60 months is 120 lg/day and for females ranging from 235 lg/day up to 445 lg/day for lactating mothers. An estimated bioconversion 6:1 ratio of b-carotene equivalents (b-CE) to vitamin A from cooked banana pulp was used (Bresnahan et al., 2012) with an estimated consumption of

cooked bananas of 300 g/day for children and 500 g/day for

women. The a- and b-carotene retention after steaming or boiling was also estimated at 70% (Mbabazi, 2015). Using these

parameters, banana fruit needed to contain b-CE levels of at least 20 lg/g dw to achieve 50%of the EAR.

There were three major technical constraints at the com-

mencement of this project that influenced the research strategy:

(i) very little information was available regarding the expression of

transgenes in bananas generally and more specifically in banana

fruit, (ii) the time from transformation to harvestable fruit ranges

from 2 to 2½ years, and (iii) it was clearly impractical to take large numbers of transgenic bananas through to fruit in the green-

house. Therefore, a large number of independent transgenic lines

were generated to enable the testing of a wide range of

promoter and transgene combinations. Initially, a single plant per

transgenic line was planted in the field with, in most instances,

between 10 and 30 transgenic lines per construct. For more

information, refer to the ‘History of the project’ section of the

Supplementary information document.

Promoter characterization

Three promoters were selected as possible candidates for

expressing PVA-related transgenes in banana fruit, and these

were characterized for levels and patterns of expression in

transgenic bananas. The promoters included the constitutive

maize polyubiquitin promoter (Ubi) and two promoters isolated

from banana, the expansin 1 promoter (Exp1) and the ACC

oxidase promoter (ACO) which were predicted to be fruit

specific. These promoters were fused to the b-glucuronidase reporter gene (uidA), the cassettes transformed into bananas

and the transgenic plants established in the field. b-glucur- onidase (GUS) protein levels were measured in the fruit pulp

using ELISA; however, this approach could not be used for leaf

or peel material because of very high background levels.

Therefore, for leaf and peel samples, MUG fluorometric assays

were used to estimate enzyme activity rather than protein

levels.

GUS activity was measured in the leaves of six independent

transgenic lines for each promoter. As expected, there were high

but variable levels of GUS activity in the leaves of all six plants

where uidA was under the control of the Ubi promoter

(Figure 1a). In the leaves of the wild-type control plant and

plants where uidA was under the control of either the Exp1 or

ACO promoter, there was undetectable to negligible GUS activity

(Figure 1b and c).

Pulp samples from the fruit of the same lines described

above were collected at 3, 6, 9 and 12 weeks post-bunch

emergence (S3, S6, S9 and S12) and also at ‘full green’ (FG),

when the bunches were harvested, and ‘full ripe’ (FR). The FG

stage is equivalent to the stage when cooking bananas are

harvested in Uganda. GUS protein was not detected in the

wild-type at any fruit development stage. In contrast, appre-

ciable but varying levels of GUS protein were detected in the

pulp of banana fruit from S3 through to FR in the six Ubi-uidA

lines (Figure 1d) confirming the constitutive nature of the Ubi

promoter. No reproducible trend in GUS protein levels across

the six lines was observed as fruit matured from S3 to FR

except that average protein accumulation was lowest at the

earliest stage, S3. In the six Exp1-uidA lines examined, no

appreciable GUS expression was detected in fruit pulp from S3

through to FG with the exception of FR fruit from lines FT258

and FT263 (Figure 1e). This indicated that the Exp1 promoter is

activated very late during fruit development. Very low levels of

GUS expression were detected in the fruit pulp of the ACO-

uidA lines from S3 to S9 (Figure 1f). With the exception of line

FT736 which peaked fourfold higher than any other line, the

overall trend was that GUS expression slowly increased from S9

through to FG and plateaued at FR. These results indicated that

the ACO promoter was activated earlier than Exp1 during fruit

ª 2016 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 15, 520–532

Pro-vitamin A biofortified Cavendish bananas in the field 521

development and more consistently at the FG and FR stages

(Figure 1e and f).

The pattern of Ubi driven GUS expression in fruit peel was

unexpectedly different from fruit pulp with low expression from

S3 to FG followed by a substantial increase at FR (Figure 1g). In

the peel of Exp1-uidA lines, the pattern of GUS expression was

similar to that observed in the fruit pulp (Figure 1h). ACO-uidA

lines had very little GUS expression in the peel up to FG; however,

there was a dramatic increase to levels higher than either the Ubi

or Exp1 lines at FR (Figure 1i).

PVA analysis: plant and ratoon crops

Domesticated bananas grow as a perennial crop. The initial plant,

the plant crop, develops a corm and a pseudostem from which

the original bunch is produced. After the bunch is harvested, the

pseudostem dies and is replaced by a second pseudostem

producing the first ratoon crop which develops from a sucker

originating from the corm. Similarly, a second ratoon crop is

produced and so on.

The same three promoters used to assess GUS expression were

used to drive the expression of three carotenoid biosynthesis

transgenes. These transgenes were the phytoene synthase 1 gene

from maize (ZmPsy1) used in Golden Rice 2, a phytoene synthase

2a gene isolated from the Fe’i banana cultivar Asupina (MtPsy2a)

(Mlalazi et al., 2012) and the bacterial phytoene desaturase gene

PaCrtI also used in Golden Rice 2. ZmPsy1 and MtPsy2a were

transformed into Cavendish banana singly and in combination

with PaCrtI.

A total of 244 transgenic lines were confirmed to contain the

respective PVA transgene(s) by PCR. Southern blot analyses

showed that the transgene copy number varied from one to more

than 10 copies (Figure S1 and S2). For each line, a single plant

was established in the field together with 50 non-transgenic

control plants. This initial randomized trial was designated Field

Trial 1 (FT-1), and the plant crop was assessed over a period of

16 months. Forty-eight transgenic lines either died or were

stunted and did not produce fruit. Therefore, fruit was harvested

from a total of 196 transgenic lines of which 153 samples were

selected for the initial plant crop fruit analysis (Table 1). Fruit was

harvested at FG, ripened to FR and sampled at both stages. The

sample used for the initial PVA level screen consisted of a single

fruit taken from the middle of the bunch from each of 153

transgenic lines together with the equivalent sample from the

fruit of 15 non-transgenic control plants. Following lyophilization

and total carotenoids extraction, each sample was analyzed by

HPLC and b-CE levels calculated. The most important outcomes

Figure 1 Analysis of promoter activity in wild-type and transgenic Cavendish banana lines. GUS activity was measured in leaf (a, b, c), and peel (g, h, i),

while GUS protein concentration was measured in pulp tissue (d, e, f). Ubi promoter (a, d, g); Exp1 promoter (b, e, h); and ACO promoter (c, f, i). WT, wild-

type FT432. S3, S6 and S9 represent 3, 6 and 9 weeks post-bunch emergence, respectively. FG, full green and FR, full ripe. Error bars: ?SD.

ª 2016 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 15, 520–532

Jean-Yves Paul et al.522

of this initial screening included: (i) there was obvious variation

between individual control plants and also variation within lines

with the same promoter-transgene combination, (ii) the highest

expressing transgenic line for each of the different transgenes or

combinations of transgenes contained higher levels of b-CE than the highest control plants except those lines containing Exp1-

PaCrtI or Ubi-PaCrtI alone or where Ubi-PaCrtI was combined with

Exp1-MtPsy2a or Exp1-ZmPsy1, (iii) in all transgenic lines and

controls, the fruit contained higher levels of a-carotene than b- carotene (Figure S3), and (iv) PaCrtI alone had little effect on fruit

PVA levels when driven by either the Exp1 promoter or the Ubi

promoter; as such, none of the single transgene PaCrtI lines were

progressed through for further analysis.

From this initial plant crop screen, 63 transgenic lines were

selected for more comprehensive analysis. The selection included

lines representing high, average and low PVA accumulation. As

preliminary data revealed considerable variation in PVA accumu-

lation across the bunch (data not shown), carotenoids were

extracted from a composite sample including equal amounts of

fruit taken from the top, middle and bottom of the bunch.

Further, although fruit samples from the selected lines were to be

analyzed across three generations (plant crop followed by first

and second ratoon crops), the trial was hit by a severe cyclone in

February 2011. As a consequence, all lines were blown over

resulting in fruit from some lines in either the first or the second

ratoon crops not being available for analysis.

For this project, the PVA levels at FG were considered more

important as cooking bananas are harvested at this stage in

Uganda. However, FR data were also collected as cooking

bananas are usually consumed over a number of days post-

harvest. The b-CE levels in FG and FR fruit from each of the selected 63 transgenic lines are presented in Table 2. Where

expression of MtPsy2a or ZmPsy1 was controlled by the Exp1

promoter, the level of b-CE in the plant crop and first ratoon increased from FG to FR in 43 of 53 samples analyzed (81%)

(Table 2). A similar increase was seen in 82% (14 of 17 samples)

of the analyzed samples where the Ubi promoter was used to

drive the expression of the same transgenes. However, when

ACO was used as a promoter, this number reduced to 50% (13

of 26 samples).

In the plant crop, there were no lines that met the target level

of 20 lg/g dw b-CE where the composite sample of three fruit per bunch was analyzed. The highest b-CE level at FG was 18.7 lg/g dw in Ubi-MtPsy2a line FT328 that subsequently died. In addition, only eight lines had PVA levels greater than 10.0 lg/g dw b-CE at FG. These were two Ubi-ZmPsy1 lines (FT287 and FT 309 with 13.4 and 11.9 lg/g dw b-CE, respectively), two Ubi- MtPsy2a lines (FT328 and FT324 with 18.7 and 11.7 lg/g dw b- CE, respectively), two ACO-MtPsy2a lines (FT504 and FT518 with

16.6 and 15.9 lg/g dw b-CE, respectively) and two ACO- ZmPsy1 + Exp1-PaCrtI lines (FT584 and FT587 with 17.1 and 11.5 lg/g dw b-CE, respectively). To investigate whether a correlation existed between transgene expression levels and

accumulation of carotenoids, the expression of the ZmPsy1 and

MtPsy2a transgenes was determined in the FG fruit of a selection

of lines using RT-PCR (Figure S4 and S5) and qRT-PCR (Figure 2).

MtPsy2a line FT246 had the highest relative expression of the

transgene followed by line FT324, FT518 and FT295 (Figure 2a).

Expression was considerably lower in the other three lines tested.

Expression of ZmPsy1 was highest in line FT584 followed by

FT309 while similar, but lower expression was seen in lines FT287,

FT467, FT475, FT479 and FT585 (Figure 2b). Lines FT187 and

FT192 had low expression.

When FG fruit from the next generation (first ratoon crop)

were analyzed, 68% (23 of 34) of samples showed an increase in

b-CE from the plant crop (Table 2). Interestingly, every line

Table 1 Data summary from the transgenic banana field trial

Promoter-transgene

Number of plants

in the field

Number of plants

harvested

Number of plants

analyzed (first cut)*

b-CE in FG fruit

(lg/g dw)

b-CE in FR fruit

(lg/g dw)

Range Average Range Average

Wild-type 50 50 15 0.6–3.8 1.5 0.8–5.8 3.1

Exp1-MtPsy2a 33 28 28 1.2–8.6 3.3 0.8–10.0 2.8

Exp1-MtPsy2a + Ubi-PaCrtI 13 8 5 0.2–2.3 1.0 1.3–3.4 2.1

Exp1-ZmPsy1 32 26 26 1.2–4.6 2.6 1.3–15.2 4.5

Exp1-ZmPsy1+ Ubi-PaCrtI 18 11 8 0.2–1.8 1.0 1.5–3.6 2.4

Exp1-ZmPsy1+ Exp1-PaCrtI 7 7 2 – – 2.1–11.0 6.5

Exp1-PaCrtI 29 26 3 – – 1.7–2.6 2

ACO-MtPsy2a 30 29 28 11.2–15.3 13.3 3.4–13.4 8.3

ACO-ZmPsy1 30 18 18 5.4–10.9 7.6 2.5–24.6 9.0

ACO-ZmPsy1+ Exp1-PaCrtI 6 4 3 10.6–25.7 17.1 7.8–16.7 13.2

Ubi-MtPsy2a 9 7 7 1.4–19.1 6.8 3.5–16.1 7.4

Ubi-ZmPsy1 10 5 5 0.3–13.6 6.1 1.0–16.1 7.9

Ubi-PaCrtI 27 27 20 – – 1.0–3.7 1.8

Total 294 246 168 – – – –

? First cut relates to an initial screening of the fruit of transgenic banana lines done by HPLC and using only a single fruit collected from the middle position of the bunch. FG, full

green and FR, full ripe.

ª 2016 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 15, 520–532

Pro-vitamin A biofortified Cavendish bananas in the field 523

Table 2 PVA carotenoid concentration in the fruit pulp of selected wild-type and transgenic Cavendish banana lines across four generations

Promoter-transgene Line #

Plant crop 1st ratoon crop 2nd ratoon crop Sucker crop

FG FR FG FR FG FR FG FR

b-carotene equivalents (lg/g dw)

Wild-type Average (n? = 6) 2.6 3.1 2.2 2.2 1.7 2.3 6.0 7.4

Exp1-MtPsy2a FT246 7.3 9.3 8.3 8.5 NA NA 18.2 19.6

FT544 3.4 4.6 NA NA NA NA NA NA

FT545 3.4 4.7 NA NA NA NA NA NA

FT342 2.8 4.2 7.4 4.6 NA NA 10.6 9.0

FT335 2.5 4.0 2.4 3.3 NA NA NA NA

FT343 2.3 3.8 2.9 3.4 NA NA NA NA

FT242 2.2 3.7 2.0 2.0 NA NA 8.6 8.0

FT233 2.0 3.3 1.9 2.6 NA NA NA NA

FT341 1.4 3.1 1.8 2.0 NA NA 9.3 8.1

Exp1-MtPsy2a + Ubi-PaCrtI FT244 1.5 2.8 2.6 1.9 NA NA NA NA

FT245 0.9 1.8 1.8 4.6 NA NA NA NA

FT220 0.7 1.4 1.7 1.9 NA NA NA NA

FT232 0.2 1.9 1.8 1.5 NA NA NA NA

Exp1-ZmPsy1 FT534 9.3 11.9 NA NA NA NA NA NA

FT536 9.1 8.9 NA NA NA NA NA NA

FT317 6.6 7.9 NA NA NA NA 11.4 15.6

FT192 4.3 6.4 NA NA NA NA 20.3 31.2

FT538 3.8 9.4 NA NA NA NA 10.9 14.8

FT187 3.5 5.3 1.8 1.8 9.5 9.9 7.0 11.0

FT311 2.9 6.1 3.1 2.5 NA NA NA NA

FT201 2.3 3.2 1.9 3.2 NA NA 9.1 15.0

FT319 1.9 3.2 3.5 2.6 NA NA NA NA

FT318 1.5 2.7 1.9 2.4 NA NA NA NA

FT210 1.3 2.8 1.2 2.5 NA NA NA NA

Exp1-ZmPsy1 + Ubi-PaCrtI FT217 1.8 2.4 1.5 2.4 NA NA NA NA

FT196 1.6 4.2 2.9 2.5 NA NA NA NA

FT207 1.5 3.5 2.4 2.7 NA NA NA NA

FT195 1.0 2.6 NA NA NA NA NA NA

FT208 0.8 2.7 2.1 2.1 NA NA NA NA

FT205 0.7 2.6 1.0 7.4 NA NA NA NA

FT213 0.4 1.8 NA NA NA NA NA NA

ACO-MtPsy2a FT504 16.6 12.0 12.1 11.7 NA NA 20.0 24.7

FT518 15.9 10.7 NA NA NA NA 23.1 35.9

FT511 9.4 7.3 5.3 6.0 NA NA 14.4 13.6

FT508 9.2 7.1 4.2 3.8 NA NA 15.6 19.0

FT498 8.9 9.9 8.6 8.4 NA NA NA NA

FT506 6.0 5.9 NA NA NA NA NA NA

FT516 4.2 7.4 NA NA NA NA NA NA

FT497 4.1 10.5 NA NA NA NA 13.4 12.9

ACO-ZmPsy1 FT492 9.4 8.5 NA NA NA NA NA NA

FT467 7.3 10.4 NA NA NA NA 10.5 13.4

FT468 NA NA 7.3 6.6 NA NA NA NA

FT475 4.3 10.5 NA NA NA NA 20.7 22.0

FT493 4.3 18.7 7.0 6.0 NA NA NA NA

FT476 2.9 5.4 4.5 4.7 NA NA NA NA

FT487 2.7 15.1 NA NA NA NA NA NA

FT479 2.7 13.6 NA NA NA NA 20.5 16.2

FT483 1.7 9.1 NA NA NA NA 18.8 16.9

ACO-ZmPsy1 + Exp1-PaCrtI FT584 17.1 11.2 NA NA NA NA 27.0 32.8

FT587 11.5 NA NA NA NA NA 21.5 22.2

FT588 7.7 10.1 NA NA NA NA NA NA

FT585 7.3 5.6 NA NA NA NA 12.2 11.5

Ubi-MtPsy2a FT328 18.7 18.9 NA NA NA NA NA NA

FT324 11.7 16.1 NA NA 26.6 33.4 55.0 50.1

FT294 6.6 9.7 13.5 9.4 13.5 12.0 29.0 25.2

FT295 5.4 6.1 6.6 6.6 4.8 4.5 10.5 12.0

FT296 4.2 6.0 5.6 4.1 12.5 13.1 NA NA

FT327 3.0 4.9 NA NA NA NA NA NA

FT330 2.5 4.9 3.6 4.0 NA NA 11.9 15.1

Ubi-ZmPsy1 FT287 13.4 15.8 NA NA 24.3 21.8 39.7 60.9

FT309 11.9 14.7 NA NA 40.4 39.3 46.9 18.5

FT298 0.7 2.5 1.1 1.3 NA NA NA NA

FT302 0.6 1.5 1.4 2.0 NA NA NA NA

FG, full green and FR, full ripe.

ª 2016 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 15, 520–532

Jean-Yves Paul et al.524

analyzed that contained either Ubi-ZmPsy1 or Ubi-MtPsy2a

showed an increased accumulation in b-CE from the plant crop. However, again, no line accumulated over the 20 lg/g dw b-CE target level. FG fruit from two lines had PVA levels above 10 lg/g dw b-CE in the ratoon crop: ACO-MtPsy2a line FT504 with 12.1 lg/g dw b-CE down from 16.6 lg/g dw b-CE in the plant crop and Ubi-MtPsy2a line FT294 with 13.5 lg/g dw b-CE up from 6.6 lg/g dw b-CE in the plant crop. Due to the impact of the 2011 cyclone, only seven lines could be assessed in the

second ratoon. FG fruit from three of these lines accumulated

above target levels of PVA. Ubi-ZmPsy1 line FT309 had FG fruit

with 40.4 lg/g dw b-CE, more than double the target, while Ubi- MtPsy2a line FT324 and Ubi-ZmPsy1 FT287 accumulated 26.6

and 24.3 lg/g dw b-CE, respectively (Table 2). Carotenoid accumulation throughout fruit development was

also monitored in selected lines from each of the single promoter-

Psy combinations in the second ratoon crop at 3, 6 and 9 weeks

post-bunch emergence as well as FG and FR (Figure 3). For the

four lines where the transgene was under the control of the Ubi

promoter, PVA levels were elevated (above 15 lg/g dw b-CE) from the earliest fruit collection time point (S3) irrespective of

whether the transgene was ZmPsy1 or MtPsy2a (Figure 3a and d).

For the two lines with the highest PVA levels, the general trend

was increasing PVA during fruit development to a maximum at

FG for Ubi-ZmPsy1 line FT309 or at FR for Ubi-MtPsy2a line

FT324. In contrast, accumulated PVA levels in the fruit of Exp1-

ZmPsy1 and Exp1-MtPsy2a lines remained below 5 lg/g dw b-CE from the emergence of the bunch all the way through to S9 (with

the exception of line FT342 which accumulated 6.6 lg/g dw b-CE at S9) followed by an increase towards maturity with a maximum

of 9.9 lg/g dw b-CE at FR in Exp1-ZmPsy1 line FT187 (Figure 3b and e). During fruit development of lines containing ACO-

ZmPsy1, PVA levels were lowest at S3 and S6 for two of the lines

but moderately higher than the Exp1-ZmPsy1 lines at those stages

(Figure 3c). PVA levels peaked for those same two lines at either

S9 or FG. In contrast, the highest PVA level in line FT476 was at

S3. For the ACO-MtPsy2a lines, again one line, FT511, had

maximum PVA accumulation at S3, while the other two lines had

maximums at S9 or FG (Figure 3f). Overall, the PVA accumulation

pattern during fruit development reflected the expression profiles

previously observed in transgenic lines where the same three

promoters were used to drive the expression of uidA (Figure 1).

The constitutive Ubi promoter provided consistent stronger

expression throughout fruit development followed by the ACO

promoter and finally Exp1.

During the plant and ratoon crops, the phenotype of each

plant was recorded at regular intervals from planting to bunch

harvest. None of the 50 wild-type control plants showed altered

phenotypes and fruit developed normally (Figure 4a–c). The presence of the PaCrtI transgene did not appear to affect

phenotype. However, three categories of altered phenotypes

were observed in the transgenic lines: stunting, ‘golden leaf’ and

‘golden bunch’. For the ‘golden leaf’ phenotype, the youngest

leaf would consistently unfurl with a bright yellow colour

(‘golden’) and progressively turn to green as it matured (Fig-

ure 4g). Fruit on the ‘golden bunch’ emerged bright orange

instead of green (Figure 4d). As the fruit matured and filled, it

progressively turned greener to a mixture of green and orange at

harvest (Figure 4e and f). Fruits with increase PVA levels displayed

a pulp ranging from deep yellow to bright orange (Figure 4h and

i). Of the original 244 transgenic lines planted in the trial, 65 had

the ‘golden leaf’ phenotype of which 29 were also stunted; 29

had the …

Read more
Applied Sciences
Architecture and Design
Biology
Business & Finance
Chemistry
Computer Science
Geography
Geology
Education
Engineering
English
Environmental science
Spanish
Government
History
Human Resource Management
Information Systems
Law
Literature
Mathematics
Nursing
Physics
Political Science
Psychology
Reading
Science
Social Science
Home
Homework Answers
Blog
Archive
Tags
Reviews
Contact
google+twitterfacebook
Copyright © 2021 SweetStudy.com