5.1 Introduction to the GMOs

18. The applicant proposes to release 25 categories of GM sugarcane with gene sequences from a total of 19 genes of interest, two marker genes and one reporter gene (Table 2). With three exceptions, each category is based upon one full or partial sequence from a gene of interest, which would be combined with different regulatory elements. Two categories would contain two marker genes with and without a reporter gene, and one category would contain a combination of five genes from other categories. Within each category, a group of expression cassettes would be made consisting of the gene(s) of interest in combination with up to five promoters, up to three terminators and up to two targeting sequences (Table 3, Table 4). Within each category, expression cassettes may be cloned into a binary vector for Agrobacterium-mediated transformation and up to two different expression vectors for transformation by microprojectile bombardment.

19. The GM sugarcane categories, with the exception of two categories containing only marker and reporter genes, would contain genes that may affect plant growth, drought tolerance, nitrogen use efficiency, sucrose accumulation, and the efficacy of cellulosic ethanol production from sugarcane biomass. Risks associated with dealings with GM sugarcane containing nine of the 19 genes of interest have been previously assessed in the RARMP for DIR 070/2006 (OGTR 2007b), however in the current application these genes are proposed to be combined with a greater variety of promoter sequences, some may be crossed with other lines, and the genes may be introduced into other sugarcane cultivars.

20. Six categories of GM sugarcane would contain partial or complete gene sequences expected to alter plant growth of the GM sugarcane plants, with expected phenotypes including decreased or increased height and decreased or increased tillering. These genes are PcGA2ox-1 derived from runner bean, HvGA20ox-1 and HvGA20ox-2 from barley, OsTB1 from rice and ShTB1 from sugarcane.

21. Three categories of GM sugarcane would contain genes expected to confer enhanced drought tolerance. One gene, OsDREB1A, is from rice and the other two genes are from a common plant and a common bacterium.

22. One category of GM sugarcane would contain the gene ZmDof1 from maize, which is expected to confer enhanced nitrogen use efficiency.

23. Seven categories of GM sugarcane would contain partial gene sequences expected to alter sucrose accumulation by modifying sucrose transport, carbohydrate metabolism or osmotic stress tolerance. The partial gene sequences are derived from a common plant species, and are incorporated into constructs designed to decrease expression of homologous genes in sugarcane.

Table 2. The genes used to alter plant growth, improve drought tolerance, improve nitrogen use efficiency, alter sucrose accumulation and improve cellulosic ethanol production from sugarcane biomass in sugarcane

GeneGenbank AccessionFunction of proteinSourceCategoriesIntended purpose
PcGA2ox-1*AJ132438Enzyme involved in degradation of active gibberellinPhaseolus coccineus
(runner bean)
3Altered plant growth
HvGA20ox-1*AY551428Enzyme involved in biosynthesis of active gibberellin Hordeum vulgare (barley)4Altered plant growth
HvGA20ox-2*UnpublishedEnzyme involved in biosynthesis of active gibberellinHordeum vulgare (barley)5Altered plant growth
OsTB1*AB088343Transcription factor regulating axillary bud growthOryza sativa (rice)6Altered plant growth
ShTB1* UnpublishedTranscription factor regulating axillary bud growthSaccharum spp. (sugarcane)7, 8Altered plant growth
WUE1 9Improved drought tolerance
WUE2 10Improved drought tolerance
OsDREB1AaAF300970Transcription factor potentially regulating drought responseOryza sativa (rice)11Improved drought tolerance
ZmDof1*D78377Transcription factor potentially affecting nitrogen useZea mays (maize)12Improved nitrogen use efficiency
SA1 CCICCICCI13, 14Altered sucrose accumulation
SA2 CCICCICCI15Altered sucrose accumulation
SA3 CCICCICCI16Altered sucrose accumulation
SA4 CCICCICCI17, 18Altered sucrose accumulation
SA5 CCICCICCI19Altered sucrose accumulation
SA6 CCICCICCI20, 25Modified plant cell wall chemical structure
SA7 CCICCICCI21, 25Production of cell wall modifying enzymes
SA8 CCICCICCI22, 25Production of cell wall modifying enzymes
SA9 CCICCICCI23, 25Production of cell wall modifying enzymes
SA10 CCICCICCI24, 25Production of cell wall modifying enzymes
Neomycin phospho-transferase II (nptII)*AAF65403Antibiotic resistanceEscherichia coli1-25Plant selectable marker
-lactamase (bla)*AJ847363Antibiotic resistanceEscherichia coli1-25 Bacterial selectable marker
uidA (GUS)*AY292368β-glucuronidaseEscherichia coli2Reporter gene
* Risks associated with dealings with GM sugarcane containing these genes were assessed previously in the RARMP for DIR 070/2006 (OGTR 2007b) , in which ShTB1 was called SoTB1.
a The applicant originally specified inclusion of a gene called OsDREB1 (accession AY196209), which was changed to OsDREB1A during the assessment process.

Table 3. The regulatory sequences used in the genetic modification of sugarcane
Regulatory SequenceGenbank AccessionFunctionSource
Ubi1S94464Constitutive promoterZea mays (maize)
P1 CCICCIHordeum vulgare (barley)
P2CCICCISorghum bicolour (sorghum)
P3CCICCIArabidopsis thaliana
P4CCICCIArabidopsis thaliana
Legumain targeting domainDQ458784Vacuolar targetingSaccharum spp. (sugarcane)
Rubisco small subunit targeting domainX04334Plastid targetingPisum sativum (pea)
nos (nopaline synthase)V00087Termination regionA. tumefaciens
ocs (octapine synthase)X00493Termination regionA. tumefaciens
tml (tumour morphology large)X00493Termination regionA. tumefaciens

24. Six categories of GM sugarcane would contain genes, derived from two species of bacteria and a common plant, that are expected to modify the plant cell wall chemical structure or cause sub-cellular accumulation of cell wall degrading enzymes. The aim of these modifications is to increase the efficiency of post-harvest processing of sugarcane biomass for cellulosic ethanol production, through improving recovery of fermentable sugars.

25. One category of GM sugarcane would contain a reporter gene (uidA) encoding an enzyme (β-glucuronidase, GUS) that enables visual identification of plant tissues in which this gene is being expressed. The reporter gene was originally derived from the common gut bacterium Escherichia coli.

26. All of the GM sugarcane categories would contain an antibiotic resistance selectable marker gene, neomycin phosphotransferase II (nptII), including one category which would contain no other plant-expressed transgenes. The nptII gene was originally derived from the common gut bacterium E. coli, and confers resistance to antibiotics such as geneticin and paromomycin on the GM plant. The nptII gene would only be used a selective marker during early stages of development of the GM plants in the laboratory.

27. The two categories of GM sugarcane containing only marker and reporter genes are for the purpose of evaluating the effects of the transformation process on the agronomic properties of sugarcane. The agronomic properties of these lines would be used as a baseline against which the applicant could measure how agronomic characteristics were changed in GM lines from other categories.

28. Additionally, expression vectors for biolistic transformation would contain the marker gene bla from the bacterium E. coli, which confers ampicillin resistance. It is expressed from a bacterial promoter that does not function in plants, so the gene is not expressed in the GM sugarcane plants. The gene would only be used to select bacteria containing the desired genes in the laboratory, prior to the production of the genetically modified plants. In constructs produced for Agrobacterium-mediated transformation, the region of DNA to be transferred would not contain bla.

29. Short regulatory sequences would be used to control expression of the genes (Table 3). These sequences are derived from plants (including maize, sugarcane and pea), a soil bacterium (Agrobacterium tumefaciens) and E. coli. Although A. tumefaciens is a plant pathogen and E. coli is a facultative human pathogen, the regulatory sequences comprise only a small part of their respective total genomes, and are not capable of causing disease.

Table 4. Characteristics of the categories of GM sugarcane

CategoryGene(s)aConstruct functionPromotersTerminatorsTargeting sequencesExpected phenotype
1*nptIIExpressionUbi1nosnoneMarker gene expression
2nptII + uidAExpressionUbi1nosnoneMarker and reporter gene expression
3PcGA2ox-1ExpressionAny from Table 3nosnoneShorter plants
4*HvGA20ox-1ExpressionAny from Table 3nosnoneTaller plants
5*HvGA20ox-2ExpressionAny from Table 3nosnoneTaller plants
6OsTB1ExpressionAny from Table 3nosnoneDecreased no. of tillers
7ShTB1ExpressionAny from Table 3nosnoneDecreased no. of tillers
8ShTB1RNAiAny from Table 3nosnoneIncreased no. of tillers
9*WUE1ExpressionAny from Table 3nosAny from Table 3 or noneImproved drought tolerance
10*WUE2ExpressionAny from Table 3nosnoneImproved drought tolerance
11*OsDREB1AExpressionAny from Table 3nosnoneImproved drought tolerance
12ZmDof1ExpressionAny from Table 3nosnoneImproved nitrogen assimilation
13SA1 (fragment 1)RNAiAny from Table 3Any from Table 3noneAltered sucrose transport
14SA1 (fragment 2)RNAiAny from Table 3Any from Table 3noneAltered sucrose transport
15SA2RNAiAny from Table 3Any from Table 3noneIncreased sugar accumulation
16SA3RNAiAny from Table 3Any from Table 3noneAltered carbohydrate partitioning
17SA4 (fragment 1)RNAiUbi1Any from Table 3noneAltered carbohydrate partitioning
18bSA4 (fragment 2)RNAiUbi1Any from Table 3noneAltered carbohydrate partitioning
19SA5RNAiAny from Table 3Any from Table 3noneAltered carbohydrate partitioning
20SA6RNAiUbi1nosnoneModified cell wall structure
21SA7ExpressionAny from Table 3nosAny from Table 3 or noneAccumulation of a cell wall modifying enzyme (no phenotype)
22SA8ExpressionAny from Table 3nosAny from Table 3 or noneAccumulation of a cell wall modifying enzyme (no phenotype)
23SA9ExpressionAny from Table 3nosAny from Table 3 or noneAccumulation of a cell wall modifying enzyme (no phenotype)
24SA10ExpressionAny from Table 3nosAny from Table 3 or noneAccumulation of a cell wall modifying enzyme (no phenotype)
25SA6 (RNAi) + SA7 + SA8 + SA9 + SA10Expression and RNAiAny from Table 3nosAny from Table 3 or noneModified cell wall chemical structure and accumulation of cell wall modifying enzymes
a The bla gene, which is not expressed in plants, would be present in all constructs for biolistic transformation with plasmid vectors. The nptII gene would be present in all constructs. All constructs except those noted as RNAi would contain the entire coding region of the selected gene. RNAi constructs would contain partial gene sequences.
b This category was not included in the original application, but was added during the assessment process.
* Selected clones from these categories are proposed to be used for crossing to non-GM sugarcane and to other GM lines selected for crossing within this release.

5.2 The introduced genes or RNAi constructs, their encoded proteins and their associated effects

30. In most cases, full gene sequences are proposed to be used for the genetic modifications. The purpose of these modifications is to introduce proteins with new functions into sugarcane, to increase expression of sugarcane genes, or to introduce proteins with similar functions to sugarcane genes which would be expressed in different ways to the endogenous genes.

31. In some instances, partial rather than complete gene sequences are proposed to be used for the genetic modifications, with the aim of decreasing expression of specific endogenous sugarcane genes. The decrease in endogenous gene expression is brought about by a mechanism known as gene silencing or RNA interference (RNAi). RNAi is a plant defence against infecting RNA viruses, and works by using RNA sequences identified by a plant as foreign to recognise matching sequences, which are then destroyed by enzymes (reviewed by Baulcombe 2004). In viral infections, the action of this mechanism means that once a sequence has been recognised as belonging to a virus, any matching sequences belonging to replicating viruses are quickly destroyed.

32. For a transgene to induce RNAi against an endogenous gene, the transcript from the transgene must mimic the structure of the double-stranded RNA viruses which naturally induce RNAi, using sequences from the gene to be silenced (the target gene) (reviewed by Waterhouse & Helliwell 2003). RNAi transgene constructs typically consist of two copies of a fragment of the target gene, arranged to give rise to a single transcript with one forward orientation copy of the target gene sequence followed by one reverse orientation copy. Because the transcript contains identical gene fragments in opposite orientations, they are complementary and naturally base-pair into a double-stranded RNA structure. The double-stranded RNA structure is recognised as being virus-like by the cellular RNAi machinery, which then cuts the transcript into fragments of 21-24 nucleotides. The RNA fragments become sequence guides for enzymes which destroy complementary RNA sequences, including any endogenous transcript with sequence closely matching the transgene. Through this pathway, there is a strong decrease in levels of endogenous transcripts highly similar to the RNAi construct. Because the transcript from the construct is destroyed in this process, no proteins are produced.

5.2.1 Genes expected to alter plant growth

33. Six categories of GM sugarcane expressing sequences from five different genes (PcGA2ox-1, HvGA20ox-1, HvGA20ox-2, OsTB1 and ShTB1) are expected to alter plant growth by affecting characteristics such as internode length and tillering. Risks that may be associated with all of these genes were assessed previously in the RARMP for DIR 070/2006 (OGTR 2007b). Background information on the effects of altered plant growth (plant architecture) on plant productivity has been given in the RARMP for DIR 070/2006 (OGTR 2007b) and will not be discussed here.

34. The genes PcGA2ox-1 (P. coccineus Gibberellin 2-oxidase-1), HvGA20ox-1 and HvGA20ox-2 (H. vulgare Gibberellin 20-oxidase-1 and -2) encode enzymes involved in gibberellin (GA) biosynthesis. PcGA2ox-1 oxidises a biologically active form of GA to an inactive form, resulting in decreased GA activity, and HvGA20ox-1 and -2 oxidise an inactive form of GA to an active form, resulting in increased GA activity (Hedden & Kamiya 1997). Biologically active GA generally stimulates internode elongation (reviewed by Raven et al. 1999). These genes would be expressed from a range of promoters in the GM sugarcane lines, resulting in changes to plant height and potentially yield.

35. OsTB1 (O. sativa Teosinte branched1) and ShTB1 ShTB1 was referred to as SoTB1 in the RARMP for DIR 070/2006 (Saccharum hybrid Teosinte branched1) encode transcription factors that may affect tillering in the GM sugarcane lines. OsTB1 and ShTB1 are homologous to the maize TB1 gene which controls branching in the maize plant (Doebley et al. 1997; Hubbard et al. 2002). Over-expression of TB1 reduces tillering in GM rice (Takeda et al. 2003; McSteen & Leyser 2005). As in DIR 070/2006, in DIR 095 it is proposed that a complete gene sequences of ShTB1 and OsTB1 would be expressed, and a partial gene sequence of ShTB1 in an RNAi construct would induce silencing of the endogenous sugarcane gene. Expression of the complete gene sequences in GM sugarcane is expected to lead to reduced tillering while RNAi is expected to result in an increase in tillering and reduction in height.

5.2.2 Genes expected to improve drought tolerance

36. Three categories of GM sugarcane expressing three different genes (WUE1, WUE2 and OsDREB1A) are expected to have enhanced drought tolerance. Background information on the effects of enhanced drought tolerance on plant productivity has been given in the RARMP for DIR 070/2006 (OGTR 2007b) and will not be discussed here. Assessment of the risks associated with dealings with GM sugarcane under DIR 070/2006 involved different genes to those expected to confer enhanced drought tolerance in the current application. The specific identities of WUE1 and WUE2 and the phenotypes resulting from their expression have been declared CCI and are not discussed further in this Section.

37. The gene OsDREB1A (O. sativa Dehydration-responsive element-binding protein 1A) from rice encodes a transcription factor belonging to the 139-member ERF family. This family is divided based upon amino acid sequence of the DNA-binding AP2 domain: there are two sub-families, the CBF/DREB subfamily (to which OsDREB1A belongs) and the ERF subfamily. The CBF/DREB subfamily is comprised of four major subgroups, with OsDREB1A belonging to subgroup III (Nakano et al. 2006). These divisions are thought to reflect changes in DNA-binding specificity which may have biological significance. While most members of this large family are unstudied, known functions are diverse. Members of CBF/DREB group III are known to have roles in abiotic stress responses. Several, including OsDREB1A, have been shown to bind to a DNA sequence known as the dehydration-responsive element (DRE), a promoter element common to genes up-regulated in response to drought, high-salt and cold stresses (Dubouzet et al. 2003). Through binding to DREs, DREB proteins can mediate broad transcriptional responses to dehydration stress. Over-expression of some DREB genes, including OsDREB1A, has been shown to result in increased drought tolerance (reviewed by Nakashima et al. 2009). For example, in Arabidopsis thaliana, strong constitutive expression of A. thaliana DREB1A was found to confer drought tolerance and cause severe growth retardation, while expression of DREB1A from a stress-responsive promoter minimised negative effects on growth while still conferring stress tolerance (Kasuga et al. 1999).

5.2.3 Genes expected to improve nitrogen use efficiency

38. Background information on the effects of enhanced nitrogen use efficiency on plant productivity has been given in the RARMP for DIR 070/2006 (OGTR 2007b) and will not be discussed here. The gene expected to confer enhanced nitrogen use efficiency on GM sugarcane lines assessed under DIR 070/2006 was the same gene as is included in the current application.

39. One category of GM sugarcane expressing a DNA-binding with one finger (Dof) transcription factor is expected to have enhanced nitrogen use efficiency. Dof transcription factors are specific to plants and have a wide range of functions which they carry out by binding to promoter elements and enhancing expression of target genes (reviewed by Yanagisawa 2002). They are also known to bind to other proteins, including bZIP transcription factors (reviewed by Yanagisawa 2004). ZmDof1 (Z. mays Dof1) is thought to regulate multiple light-responsive genes involved in synthesis of organic acids from fixed carbon, a process which gives rise to the carbon skeleton from which amino acids are synthesised following nitrogen assimilation (reviewed by Yanagisawa 2004). A study in A. thaliana showed that constitutive expression of ZmDof1 resulted in up-regulation of genes involved in carbon skeleton production (Yanagisawa et al. 2004). In these plants an increase in amino acid content was observed, along with enhanced nitrogen assimilation under normal conditions and increased growth under low nitrogen conditions. Expression of ZmDof1 in GM sugarcane may result in similar improvements to nitrogen assimilation, which may improve plant growth when nitrogen fertiliser inputs are reduced.

5.2.4 Genes expected to alter sucrose accumulation

40. Sugar is manufactured during photosynthesis in leaves, mainly in the form of sucrose, and is transported in the phloem to other parts of the plant for growth and, in sugarcane, to the stem for storage. Large amounts of sucrose are deposited in the sugarcane stem in both the vacuoles of storage parenchyma cells and the apoplast (cell wall and intercellular spaces) surrounding these cells (reviewed by Braun & Slewinski 2009). Sugarcane is able to store significant amounts of sucrose and a mature stem can accumulate up to approximately 17% of its fresh weight as sucrose (Bull & Glasziou 1963).

41. Increasing the sucrose content of sugarcane is a major objective of most sugarcane improvement programmes. Grof and Campbell (2001) identified the four rate-limiting steps of sucrose accumulation as the leaf reactions which produce sucrose, the rate of phloem loading for transport to the stem, the rate of transport into storage parenchyma and the rate of sucrose remobilisation for vegetative growth. Genes involved in aspects of these processes have been identified, including genes encoding proteins involved in sucrose synthesis, cleavage and transport, and carbon partitioning and storage (for example see Lakshmanan et al. 2005; Moore 2005; Casu et al. 2005). Approaches to altering rate limiting steps to increase accumulation of sucrose and other sugars include the use of genetic modification. For example, expression of a bacterial sucrose isomerase in storage parenchyma cells has been shown to result in conversion of sucrose to an isomer which is not metabolised by the plant, leading to strong increases in sugar accumulation (Wu & Birch 2007). It is thought that sugarcane is capable of accumulating sucrose to more than 25% of fresh weight, should rate-limiting steps be overcome (reviewed by Grof & Campbell 2001)

42. In the current application seven categories of GM sugarcane contain modifications which may lead to altered sucrose accumulation, all accomplished by use of RNAi constructs designed to decrease expression of sugarcane genes. The specific identities of these genes and the phenotypes resulting from their silencing have been declared CCI and are not discussed further in this Section.

5.2.5 Genes expected to improve cellulosic ethanol production from sugarcane biomass

43. Fermentation of sugars produces ethanol, a biofuel of value as a substitute for petroleum. The vast majority of current ethanol production from plant crops comes from the fermentation of starch and sugar, sourced predominantly from maize grain and sugarcane, respectively. Although cellulose is the most abundant plant carbohydrate available, it is not used for ethanol production because is significantly more expensive to produce ethanol from cellulose than starch or sugar. A major focus of current biofuel research is reducing costs and improving methods of cellulosic ethanol production, so as to enable production from a wider range of feedstock crops and avoid the use of food crops (reviewed by Sticklen 2008). If cellulosic ethanol production were to become feasible, a broad range of waste biomass remaining after crop harvest could be used to produce ethanol, including sugarcane stems from which cane juice has been extracted.

44. The major impediment to commercial cellulosic ethanol production is the expense of current production methods, in particular, the cell-wall hydrolysis enzymes needed for conversion of cellulose (a complex carbohydrate) to fermentable sugars such as glucose. Such enzymes are typically produced from fungi and bacteria in bioreactors. Current research into the production of cellulases in planta aims to decrease or eliminate the need to add enzymes during biomass processing (reviewed by Sticklen 2008). The current application includes five categories of GM sugarcane expressing four bacterial cellulase genes, either individually (categories 21-24, Table 4) or in combination (category 25, Table 4). Also included is a gene fragment derived from a common crop plant for modification of plant cell wall structure, also for the aim of improving cellulosic ethanol production from sugarcane biomass (individually in category 20 and combined with bacterial cellulase enzymes in category 25). The specific identities of these genes and the phenotypes expected to result from their expression have been declared CCI and are not discussed further in this Section.

5.2.6 Toxicity/allergenicity of the end products associated with the introduced genes or RNAi constructs

45. Risks associated with dealings with GM sugarcane containing the genes expected to give rise to altered plant growth and enhanced nitrogen use efficiency were previously assessed in the RARMP for DIR 070/2006, which concluded that there was no evidence in published scientific literature to suggest that they may be toxic or allergenic when expressed in GM sugarcane (OGTR 2007b). These genes were isolated from runner bean, barley, rice and sugarcane and homologues of all of the encoded proteins occur naturally in a range of organisms, including plants widely consumed by people and animals. On this basis humans and other organisms have a long history of exposure to these genes and their expressed proteins.

46. The WUE1 and OsDREB1A genes, which are expected to improve drought tolerance, were isolated from a common food plant and rice, respectively (Table 2), and as such humans and other organisms have a long history of exposure to these genes and their expressed proteins. WUE2 is derived from a soil bacterium, and its product is not known to be toxic or allergenic.

47. The sequences expected to improve cellulosic ethanol production from sugarcane biomass through accumulation of cellulolytic enzymes are derived from two non-pathogenic bacteria.

48. All sequences expected to alter sucrose accumulation in the GM sugarcane lines are expressed from RNAi constructs designed to reduce the levels of endogenous sugarcane transcripts, and no new proteins are expected to be produced. The secondary effects of the RNAi constructs are expected to be modifications to sucrose transport, accumulation and metabolism. It is not anticipated that such changes would alter the allergenicity or toxicity of sugarcane on the basis that sucrose is not toxic unless consumed in large quantities. The oral dose of sucrose required to kill 50% of tested rats is 29 grams per kilogram of body weight (Boyd et al. 1965).

49. A comprehensive search of the scientific literature yielded no information to suggest that any of the encoded proteins are toxic or allergenic to people, or toxic to other organisms.

50. No studies on the toxicity or allergenicity of the GM sugarcane lines and their products have been undertaken to date as the proposed trial is at an early stage. Such studies would have to be conducted if approval was sought for the GMOs or their products to be considered for human consumption in Australia.

5.2.7 The plant antibiotic resistance marker gene (nptII) and the encoded protein

51. The nptII gene is used extensively as a selectable marker in the production of GM plants (Miki & McHugh 2004). As discussed in previous DIR RARMPs, and in more detail in the RARMPs for DIR 070/2006 and DIR 074/2007 or by contacting the OGTR, regulatory agencies in Australia and in other countries have assessed the use of the nptII gene in GMOs as not posing a risk to human or animal health or to the environment. The most recent international evaluation of nptII in terms of human safety was by the European Food Safety Authority, which concluded that the use of the nptII gene as a selectable marker in GM plants (and derived food or feed) does not pose a risk to human or animal health or to the environment (EFSA 2007).

5.2.8 The bacterial antibiotic resistance marker gene (bla) and the encoded protein

52. Some of the GM sugarcane lines in the proposed release will contain the β-lactamase (bla, also known as amp) antibiotic resistance marker gene. The bla gene is derived from E. coli (Spanu et al. 2002) and encodes the β-lactamase enzyme, which confers ampicillin resistance.

53. The β-lactamase enzyme is widespread in the environment and in food. Naturally occurring ampicillin-resistant microorganisms have been found in mammalian digestive systems (Spanu et al. 2002). The bla gene was originally isolated from antibiotic resistant strains of E. coli found in hospital patients.

54. A number of GM crops containing the bla gene have been approved for limited and controlled release in Australia (eg papaya in DIR 026/2002, sugarcane in DIRs 051/2004, 070/2006 and 078/2007, and rice in DIR 052/2004). No adverse effects on humans, animals or the environment have been reported from these releases.

5.2.9 Reporter gene (uidA) and its encoded protein (GUS)

55. The uidA gene encodes the enzyme β-glucuronidase (GUS), which is derived from the common gut bacterium E. coli. The GUS protein is a monomer with a molecular weight of 68 kDa, and the GUS enzyme is active as a tetramer. GUS catalyses the hydrolysis of β-glucuronides and, less efficiently, some β-galacturonides. E. coli lives in the digestive tract of vertebrates, including humans (Jefferson et al. 1986), and the GUS enzyme enables it to metabolise β-glucuronides as a main source of carbon and energy.

56. The uidA gene is the most widely used reporter gene in GM plants (Miki & McHugh 2004) as it allows GM tissues to be identified using a simple visual assay. A number of GM crops containing the uidA gene have been approved for limited and controlled release in Australia (eg papaya in DIR 026/2002, pineapple in DIR 028/2002, grapevine in DIR 031/2002, rice in DIR 052/2004, sugarcane in DIR 070/2006 and cotton in DIR 074/2007). No adverse effects on humans, animals or the environment have been reported from these releases. The US EPA does not consider the GUS protein to be toxic and has approved its exemption from the requirements to establish tolerance levels (EPA 2001). Safety assessments by FSANZ of GM foods containing the uidA gene have concluded that its presence poses a negligible risk to human health and safety (FSANZ 2002; for example see FSANZ 2003).

5.3 The regulatory sequences

5.3.1 Regulatory sequences for expression of the introduced genes or RNAi constructs

57. Promoters are DNA sequences that are required in order to allow RNA polymerase to bind and initiate correct transcription. Five promoters are proposed to be used in the constructs (Table 3). The maize Ubiquitin1 (Ubi1) promoter is a constitutive promoter that is widely used in plant genetic modification (Christensen et al. 1992). Four other promoters would be used to obtain greater specificity of expression, and in each case the promoter sequence would be followed by the maize ubiquitin1 intron to enhance gene expression. The specific identities of these genes have been declared CCI and are not discussed further in this Section.

58. Also required for gene expression in plants is an mRNA termination region, including a polyadenylation signal. The mRNA termination regions used in the GM sugarcane lines would be derived from three A. tumefaciens genes: nopaline synthase (nos), octopine synthase (ocs) and tumour morphology large (tml). Each of these sequences has been used in a wide variety constructs for plant genetic modification, nos being very commonly used (Reiting et al. 2007). Although A. tumefaciens is a plant pathogen, the regulatory sequences comprise only a small part of its total genome, and are not capable of causing disease.

59. In RNAi constructs, separation of the inverted-repeat arms with a spliceable intron has been shown to increase the effectiveness of silencing (Smith et al. 2002). RNAi constructs for use in GM sugarcane are to be produced in the vector pStarling, in which fragments of the silencing target gene are cloned on either side of a barley Cereal cyst nematode resistance (cre) gene intron.

5.3.2 Regulatory sequences for the expression of the nptII, uidA and bla genes

60. All of the GM sugarcane lines in the proposed release are to include the nptII plant selectable marker. In lines generated by biolistic transformation, the construct containing the gene(s) of interest would be co-bombarded with a plasmid containing nptII under the control of the Z. mays Ubi1 promoter and the A. tumefaciens nos terminator. In lines generated by Agrobacterium-mediated transformation, the nptII gene under the control of the same regulatory elements would be incorporated in the transfer DNA (T-DNA) region of the binary vector containing the gene of interest (see below).

61. The uidA marker gene will also be under the control of the Z. mays Ubi1 promoter and the A. tumefaciens nos terminator.

62. The bla gene in the GM sugarcane lines will be under the control of its own bacterial promoter and terminator from E. coli and therefore would not be expressed in the GM sugarcane plants. The gene would be used in the laboratory prior to the production of the GM sugarcane lines.

5.3.3 Sequences for subcellular targeting of products of the introduced genes

63. The applicant proposes to use targeting sequences in some of the constructs (Table 3) to direct protein products into specific cellular compartments. Targeting sequences (also known as transit peptides) are amino acid motifs located at one end of a protein which contain information for protein targeting and transport (Buchanan et al. 2000). Specific targeting motifs have been found to target proteins to various organelles. This is a general approach used to increase protein production (Benchabane et al. 2008). For example, subcellular targeting has been successfully used to obtain high levels of expression of methylmercury lyase in plants (Bizily et al. 2003). Targeting the methylmercury lyase to either the endoplasmic reticulum or the cell wall resulted in GM plants with a much higher resistance to organic mercury than those GM plants in which the enzyme was expressed without a targeting sequence. The two targeting sequences to be used are the sugarcane legumain vacuole-targeting sequence and the Pisum sativum ribulose-1’5’-bisphosphate carboxylase/oxygenase (Rubisco) small subunit chloroplast-targeting domain. Depending upon the targeting sequence, the applicant proposes they would be fused to either the 5’ or 3’ end of the gene of interest (each targeting sequence would not be trialled at both ends of the gene of interest).

64. Jackson et al. (2007) used a bioinformatic approach to identify vacuolar proteins of sugarcane, and identified a legumain gene containing a vacuolar targeting motif. Legumains are asparagine-specific cysteine proteinases that, in plants, function in vacuolar compartments to process and degrade proteins (Muntz 2007). Legumains are also known as vacuole processing enzymes (VPEs). Saccharum officinarum VPE-1 is a legumain homologue with a conserved motif within its N-terminal propeptide that directs the protein into the lytic vacuole. The minimal vacuolar targeting motif, consisting of five amino acids, has been shown to direct GFP to the lytic vacuole of sugarcane when expressed as a translational fusion at the N-terminus of GFP (Jackson et al. 2007). The primary function of lytic vacuoles is amino acid recycling (Muntz 2007), and so Jackson et al. (2007) speculated that protein modifications may be necessary for avoiding degradation of proteins of interest targeted to the lytic vacuole. Vacuolar targeting has previously been successfully used to accumulate significant amounts of various transgene products (Benchabane et al. 2008).

65. The Rubisco small subunit (RbcS) is a nuclear-encoded protein which is localised to the chloroplast. Chloroplasts are a type of plastid, a group of organelles specialised for photosynthesis, storage and biosynthesis. RbcS is typical of plastid targeted proteins in that its targeting motif is an N-terminal extension (Ko et al. 2006). The RbcS targeting motif from a variety of plants has been used in a wide variety of applications to target transgene products to plastids. The P. sativum RbcS plastid targeting sequence has previously been successfully used in sugarcane to target three enzymes for polyhydroxybutyrate production to plastids (Petrasovits et al. 2007).

5.4 Method of genetic modification

66. The applicant proposes to use biolistic and Agrobacterium-mediated transformation methods to generate the GM sugarcane lines in the proposed release.

67. The biolistic transformation methods used would be based upon published methods (Bower & Birch 1992; Bower et al. 1996). Briefly, sugarcane embryogenic tissue is bombarded with tungsten particles coated with the DNA to be introduced. In the current application, this consists of a marker plasmid carrying nptII and a plasmid carrying an expression cassette for the sequences of interest. The applicant plans to perform biolistic transformation using either circular plasmid DNA or linearised plasmid DNA from which backbone sequences (including bla) have been removed. After bombardment, transformed sugarcane cells are tissue cultured under selection (on the basis of geneticin resistance conferred by the nptII gene), and regenerated into plantlets. The biolistic transformation method has been extensively used and is discussed in previous RARMPs including DIR 051/2004 and DIR 077/2007 (OGTR 2005; OGTR 2008a).

68. The applicant also proposes to generate GM sugarcane lines using Agrobacterium-mediated transformation. A. tumefaciens is a common gram-negative soil bacterium that causes crown gall disease in a wide variety of plants (Van Larebeke et al. 1974). Plants can be genetically modified by the transfer of DNA (transfer-DNA or T-DNA, located between specific border sequences on a resident plasmid) from A. tumefaciens through the mediation of genes from the virulence region of tumour-inducing plasmids. In the current application, expression cassettes carrying the sequences of interest would be introduced into the T-DNA region of binary vectors also carrying an nptII marker gene. Briefly, sugarcane callus would be treated with a culture of A. tumefaciens carrying a binary vector, then transformed cells would be regenerated to plantlets on media containing the selective agents paramomycin (resistance to which is conferred by nptII) and timentin (which suppresses A. tumefaciens). The applicant plans to PCR test DNA extracted from leaves of regenerated plantlets for the presence of Agrobacterium, to ensure it does not persist in plantlets to be transferred to the field.

69. There has been concern in a recent publication that transfer of A. tumefaciens chromosomal DNA to the plant host may, in 0.4% of cases, accompany the integration of A. tumefaciens DNA flanked by the T-DNA borders (Ulker et al. 2008). However, the likelihood of A. tumefaciens chromosomal DNA having an influence on any resulting GM plants is regarded as small given the low likelihood of plants possessing the DNA segments necessary for expression of the A. tumefaciens genes.

70. Each GM sugarcane line generated from an independent biolistic or Agrobacterium-mediated transformation event is expected to have the transgenes located at different sites in the sugarcane genome. Both biolistic and Agrobacterium-mediated transformation have been widely used in Australia and overseas for introducing new genes into plants and are not known to cause any adverse effects on human health and safety or the environment.

5.5 Characterisation of the GMOs

5.5.1 Stability and molecular characterisation

71. The applicant states that all genes to be introduced into sugarcane have been sequenced. Molecular characterisation of the GM sugarcane lines has not been carried out and stability of the genetic modifications is also unknown, as the project is in a very early stage. The applicant proposes to plant individual, uncharacterised, primary transformant plantlets to the field following screening to confirm the GM nature of plantlets.

72. Previous studies have established that genes integrated into the sugarcane genome by particle bombardment are stable (Umbeck et al. 1989; Perlak et al. 1990; Harrison et al. 2001). To demonstrate the inheritance and expression of the introduced genes, Harrison et al. (2001) crossed GM sugarcane lines with a non-GM sugarcane line. The proportion of progeny expressing the introduced gene was 1:1 indicating that the introduced genes were inherited in the normal Mendelian fashion.

73. Genes introduced into other GM sugarcane lines have also been found to be stably inherited and expressed in clones propagated via stem cuttings (Hansom et al. 1999). Sugarcane is normally propagated asexually by stem cuttings.

5.5.2 Characterisation of the phenotypes of the GM sugarcane lines

74. The proposed trial is for the purpose of characterising the phenotypes of the GM sugarcane lines in the field, and they have not undergone any prior characterisation. The applicant states that this is necessary because of limited PC2 glasshouse space and variable results of glasshouse trials.

75. Licence DIR 070/2006 authorised BSES to release GM sugarcane lines carrying some of the same genes and partial gene sequences as are described in the current application (Table 2, Table 4 categories 1-8 and 12). In DIR 070/2006 all genes and partial gene sequences were under the control of the constitutive maize Ubi1 promoter. In the current application, with the exception of nptII and uidA, all of the sequences common to DIR 070/2006 are proposed to be expressed from the maize Ubi1 promoter and four other promoter sequences thought to give more specific expression. Bearing this in mind, phenotypes reported for GM sugarcane lines released under DIR 070/2006, discussed below, are likely to bear similarities to those expected for related GM sugarcane lines in the current application. Phenotypes discussed below are based upon data provided by the applicant, which generally showed that a range of phenotypes occur for each construct. They are based upon comparisons to NPTII-expressing GM control lines, and are based upon the median of between two and 16 independent lines, each measured at least twice, from plants grown at BSES Woodford in one season. Comparison of the GM sugarcane lines to NPTII control lines is used instead of comparison to non-GM sugarcane as all plants which have been transformed show substantial decreases in weight, stem diameter and cane yield. In order to determine the effect of each specific genetic modification on sugarcane phenotype, the GM lines must be compared to other plants which have undergone transformation and regeneration in tissue culture.

GM sugarcane lines with modified plant growth

76. Lines expressing HvGA20ox-1 and -2 from the Ubi1 promoter showed a median increase in plant height of approximately 50% and a median decrease in stalk number of approximately 40-50%, and no large changes in stalk diameter. In the HvGA20ox-1 and -2 lines these changes resulted in cane yield being reduced by 25% and 50%, respectively, and sugar yield being reduced by 10% and 25%, respectively. The range of phenotypes observed for HvGA20ox-2 lines was particularly variable. The applicant has stated that expression of both HvGA20ox genes was only expected to affect plant height, with the effects on tillering being unexpected. The applicant has also stated that it is possible that buds on the stem nodes of these lines may possibly mature more quickly.

77. Lines expressing PcGA2-ox from the Ubi1 promoter generally display opposite phenotypes to the HvGA20ox lines: height was reduced by approximately 35%, stalk number was increased by approximately 30%, and stem diameter showed no large changes. As a result of these changes reductions in cane yield of approximately 45% and sugar yield of approximately 25% were observed. Similarly to the HvGA20ox-2 lines, phenotypic variability between lines was high. The applicant states that plants expressing PcGA2ox may be more resistant to lodging (falling over, typically due to heavy rain or wind).

78. Lines expressing OsTB1 and ShTB1 from the Ubi1 promoter showed little difference to control plants in height, stalk diameter, cane yield and sugar yield. OsTB1 lines showed an approximately 25% decrease in stalk number. In lines expressing RNAi constructs against ShTB1 from the Ubi1 promoter there was little effect upon plant height, an increase in median stalk number of approximately 20% and a small decrease in stem diameter of approximately 10%, with very high variability being observed between lines. These changes resulted in an approximately 35% reduction in cane yield, with sugar yield not being substantially changed.

GM sugarcane lines with enhanced nitrogen use efficiency

79. In lines expressing ZmDof1 from the Ubi1 promoter plant height was increased by approximately 20%, while small decreases in stalk number and stalk diameter were observed. As a result of these changes an approximately 20% increase in cane yield was observed, while CCS was unchanged from the control lines.