Glucosinolate and Amino Acid Biosynthesis in Arabidopsis

The Arabidopsis Col-0 MAM-Like Gene Family

Within this paper, we will refer to the four Col-0 members of the gene family as MAM1 (At5g23020), MAML (At5g23010), MAML-3 (At1g74040), and MAML-4 (At1g18500). The predicted protein for each MAM synthase contains a LeuA domain and a chloroplast leader-peptide as predicted by TargetP, P > 0.9 (Fig. 1c). MAM1 and MAML form a subgroup of the Arabidopsis MAM synthase family sharing 78% amino acid identity, while MAML-3 and MAML-4 form another subgroup sharing 90% identity (Fig. 1c).

Heterologous Expression of MAM Synthases in a ΔLeuA (IPMS-Null) E. coli Mutant

Expression constructs were designed to heterologously express the predicted mature peptide encoded by each MAM-like Arabidopsis gene, with the addition of an N-terminal 6× His tag. MAM1, MAML, MAML-3, and MAML-4 lacking their predicted leader-peptide sequences were amplified from total Arabidopsis cDNA and cloned into the E. coli expression vector pQE-30 to give pQE-MAM1, pQE-MAML, pQE-MAML-3, and pQE-MAML-4. The E. coli strain CV512 has a nonfunctional IPMS (ΔLeuA), rendering it unable to grow on media lacking Leu (Somers et al., 1973). Proteins of the predicted sizes were produced upon induction of CV512 containing either pQE-MAM1, MAML, MAML-3, or MAML-4 (Fig. 2a). Only CV512 producing the mature MAML-3 protein was able grow on minimal media lacking Leu (Fig. 2, b and c).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Complementation analysis of the E. coli Leu auxotroph CV512 expressing the empty vector control pQE-30 (0), pQE-MAM1 (1), pQE-MAML (2), pQE-MAML-3 (3), and pQE-MAML-4 (4). a, Western blot of total protein extracts from CV512 containing each construct following 2 h induction with 0.1 mm IPTG at 37°C. Peptide sizes were determined from a Coomassie gel run in parallel: (1) and (2) approximately 52 kD, (3) approximately 65 kD, and (4) approximately 64 kD. b, Growth curves of CV512 containing each construct in liquid minimal media supplemented with 0.02 mm IPTG and incubated at 37°C. Each point represents the average of three independent growth curve experiments. c, Growth of CV512 containing each construct on solid minimal media (left) + 0.1 mm IPTG (center) or + 30 mm Leu (right).

Characterization of MAM-Like Insertion Lines

As neither MAML nor MAML-4 exhibited IPMS activity, they may be involved in GSL biosynthesis. We obtained homozygous knockout (KO) lines for MAML and MAML-4. MAML::En1 (Col-0 background) contains a single, stable En-1 insertion in the 1st exon of MAML and Garlic1175 (Col-0 background) contains a single T-DNA insertion in the 10th intron of MAML-4 (Fig. 3a). These two lines are referred to as the MAML KO and MAML-4 KO lines, respectively. Reverse transcription (RT)-PCR analysis of MAML KO cDNA showed that no MAML transcript was produced (data not shown), while analysis of MAML-4 KO cDNA showed that chimeric MAML-4::T-DNA fusion transcripts were produced. However, the chimeric transcript in MAML-4 KO was present at considerably reduced levels compared to the normal transcript in Col-0 (Fig. 3b). No compensatory increase in expression was observed for the other MAM synthase genes. The growth and appearance of the MAML KO line did not differ significantly from the wild-type controls. However, the MAML-4 KO line showed a 20% reduction in germination compared to Col-0. Subsequent growth appeared normal.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Two MAM KO lines. a, Position of the En-1 transposon insertion in MAML and T-DNA insertion in MAML-4. In each case the complete gene structure is shown, white boxes denote exons while black boxes denote untranslated regions. b, Semiquantitative RT-PCR analysis of MAM gene expression in the MAML-4 KO line compared to the wild-type Col-0. The PCR product is a 539 bp section of exon 1. In each case 18 cycles of PCR amplification was used, and analysis was repeated three times to ensure reproducibility. MAM1 expression in Col-0 could be detected after 40 rounds of PCR. The constitutively expressed housekeeping gene APT was used as an mRNA loading control.

MAML Is Required for Long Chain GSLs

GSLs were analyzed in the seeds and leaves of the MAML KO line. There was a complete absence of 6C, 7C, and 8C GSLs (Fig. 4, a and b). Despite the loss of these compounds, the total level of Met-derived GSLs in the knockout and controls was not significantly different (Table II), due to enhanced levels of 4C GSLs. To test whether these alterations in GSL composition were controlled by a nonfunctional MAML allele, both the MAML KO line and wild-type Col-0 were transformed with vector pTKC28, containing a wild-type genomic copy of the MAML gene with the endogenous promoter replaced by the cauliflower mosaic virus (CAMV) 35S promoter. Production of 6C, 7C, and 8C GSLs was restored in the MAML KO line and enhanced in wild-type Col-0 (Fig. 4c; Table II). Segregation analysis of T₂ transformants and subsequent analysis of T₃ families confirmed that the enhanced levels of long chain GSLs in transgenic Col-0 was due to the presence of the transgene (Fig. 4). The total level of aliphatic GSLs was the same in the seeds of T₃ plants whether they possessed the transgene or had lost it through segregation (70.2 ± 1.31 μmol g⁻¹, n = 15 and 75.9 ± 4.79 μmol g⁻¹, n = 5, respectively), and also in leaves (20.5 ± 1.95 μmol g⁻¹ and 17.4 ± 2.41 μmol g⁻¹, respectively). However, the ratio of chain lengths varied in both tissues; those plants that contained the transgene had higher levels of long chain GSLs and 3C GSLs, but lower levels of 4C GSLs compared to wild-type plants and outsegregants in both seeds and leaves (Fig. 4).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

The MAML KO lacks 6C, 7C, and 8C GSLs. Chromatograms of desulphoglucosinolates extracted from the seeds of (a) wild-type Col-0, (b) the MAML KO line, and (c) the MAML KO line transformed with a 35S driven copy of MAML. The 6C, 7C, and 8C GSLs are indicated in bold; none were detected in the MAML KO. See Table I for peak number identification. IND, indole GSLs (derived from Trp); IS, internal standard.

Overexpression of MAML Results in the Synthesis of Novel Amino Acids

As acetyl CoA-oxo acid condensation is also an important component of amino acid biosynthesis, we investigated the amino acid content of 35S::MAML. Soluble amino acids were extracted from 9-d-old seedlings of 35S::MAML and derivatized with the AccQ Tag reagent. Amino acids were separated and detected by liquid chromatography-fluorescence detection (LC-FLD), and quantified using standards. All amino acids identified in 35S::MAML seedlings were at equivalent levels to wild type (Col-0) with the exception of Tyr. Soluble Tyr decreased more than 2-fold in 35S::MAML seedlings (37 ± 2.5 pmol mg⁻¹ in 35S::MAML and 100 ± 7.4 pmol mg⁻¹ in Col-0, n = 4, P < 0.005; Fig. 5). Furthermore, two new peaks, not seen in the wild type, were observed in the LC-FLD chromatograms of 35S::MAML extracts (Fig. 6a). These compounds were investigated by liquid chromatography-mass spectrometry (LC-MS) in derivatized 35S::MAML extracts. The most abundant ion in both the new peaks had a mass of 316, which corresponds to a prederivatization mass of 145 (Fig. 6b). The second of the two peaks has an identical retention time to derivatized homoleucine (11.6 min; Fig. 6, a and b).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Overexpression of MAML enhances the ratio of long chain to short chain aliphatic GSLs. a, GSLs in seeds of Col-0 plants homozygous for 35S::MAML and outsegregants from the same initial primary transformation event that have lost 35S::MAML. The total level of Met-derived GSLs between the two classes are not significantly different (70.2 ± 1.31 and 75.9 ± 4.79 μmol g⁻¹, respectively). b, GSLs in rosette leaves of Col-0 plants homozygous for 35S::MAML and outsegregants from the same initial primary transformation event that have lost 35S::MAML. The total level of Met-derived GSLs between the two classes is not significantly different (20.5 ± 1.95 and 17.4 ± 2.41 μmol g⁻¹, respectively).

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

Plants overexpressing MAML contain novel amino acids. a, UV-absorbance chromatograms of derivatized amino acid extracts from Col-0 seedlings, 35S::MAML seedlings, and a homoleucine standard, separated using gradient 1. There are two new peaks at 10.7 and 11.4 min in 35S::MAML seedlings. b, Single ion chromatograms at 316 amu in the same region. c, In the same extracts there is a new 330 amu peak at 32.4 min in 35S::MAML seedlings, separated using gradient 2. d, In underivatized extracts the two new compounds from 35S::MAML coelute and (e) have the same ionization spectra as homoleucine.

To obtain ionization spectra for the novel amine compounds an LC-MS/MS method was developed for underivatized extracts. Two novel peaks of mass 146 (i.e. M+H⁺) coeluted between 6.2 and 6.6 min in 35S::MAML extracts (Fig. 6d). Homoleucine also eluted at 6.5 min. Ionization spectra were obtained for both novel peaks and the authentic standard (Fig. 6e). All three mass spectra were identical, with a major transition from 146 to 100 amu. This transition is likely to be due to the loss of an HCOOH group. In summary, the underivatized compound eluting at 6.5 min is homoleucine. The earlier peak represents an isomer of homoleucine and is almost certainly homo-Ile. Further analysis suggested the occurrence of a further novel amino acid in 35S::MAML with derivatized M⁺H⁺ mass of 330, which is consistent with dihomoleucine and/or dihomoisoleucine (Fig. 6c).

MAML-4 KO Plants Have Normal GSLs but Perturbed Amino Acid Content

MAML-4 KO plants had similar GSLs to their wild-type control (data not shown), but had perturbed amino acid content. There were highly significant (P < 0.005) increases in His, Val, Asn, and Gln when compared to wild-type Col-0, and less significant decreases in Leu (P = 0.057; Fig. 7). The less significant decrease in Leu was due to just one of the four replicates that had similar levels to wild-type Col-0 (MAML-4 KO: 26.6, 27.5, 18.2, and 36.7 pmol mg⁻¹; Col-0: 40.0, 38.0, 38.0, and 32.0 pmol mg⁻¹). If this anomalous result in excluded, the level of probability of a decrease in Leu in MAM4-L KO compared to wild type decreases to P = 0.01.

• Download figure
• Open in new tab
• Download powerpoint

Figure 7.

The MAML-4 KO has significantly altered soluble amino acid composition. Mean soluble amino acid content of wild-type Col-0 (dark gray box) and MAML-4 KO (light gray box) 9-d-old seedlings. SE bars are present for each point (n = 4). Stars indicate changes that are highly significant (P < 0.005). Cys could not be reliably resolved, and Trp was below the detection threshold. Standard amino acid abbreviations are used. aab, α-amino-butyrate.

Bioinformatic Analyses

The CD search at the National Center for Biotechnology Information (Altschul et al., 1997; www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) was used to identify strongly conserved LeuA motifs in the predicted proteins of each MAM synthase. TargetP (Emanuelsson et al., 2000; www.cbs.dtu.dk/services/TargetP/) was used to identify likely chloroplast target peptides (P > 0.9) and to predict their cleavage sites: MAM1 and MAML after 49 amino acids, MAML-3 after 46 amino acids, and MAML-4 after 57 amino acids. ClustalW at EBI (www.ebi.ac.uk/clustalw/) was used to perform identity comparisons between the four Arabidopsis MAM proteins (lacking their degenerate target peptides) and the Escherichia coli LeuA (National Center for Biotechnology Information accession no. AAC73185).

Plant Growth

Plants were routinely grown in Arabidopsis mix (2 parts Levington's M3 potting compost to 1 part grit/sand) under standard glasshouse conditions at approximately 20°C. For aseptic growth, seeds were surface-sterilized and plated on growth medium (1× Murashige and Skoog salts plus vitamins [Duchefa, Haarlem, the Netherlands], 1% Suc, 0.8% agar [BACTOAGAR, Fisher Chemicals, Loughorough, UK], and 2.5 mm MES, pH 5.7). The seeds were stratified for 2 d at 4°C in the dark before germination in a growth room (16 h light/8 h dark, 20°C).

RNA Extraction and cDNA Synthesis

Leaf tissue samples were ground in liquid nitrogen, and total RNA was extracted using the RNeasy plant mini kit (Qiagen, Crawley, UK) and eluted in 40 μL of diethyl pyrocarbonate treated water. Contaminating DNA was removed by DNase treatment using the Ambion DNA-free kit (Huntingdon, UK). Five micrograms of total RNA was used to make first strand cDNA using SuperScript II (Invitrogen, Paisley, UK) in a 20-μL reaction with oligo(dT) primers according to the manufacturer's instructions. The completed reaction was diluted 50-fold, and in subsequent PCR 1 μL of the dilution was used per 10 μL of reaction mix.

cDNA Cloning

MAM1, MAML, MAML-3, and MAML-4 lacking their predicted leader-peptide sequence were PCR amplified from Col-0 leaf cDNA in a 50-μL reaction containing 5 μL of cDNA dilution and 2 units of PfuUltra (Stratagene, Amsterdam), 1× supplied buffer, 0.3 μm each primer, and 0.2 mm dNTPs. An initial denaturation step of 96°C for 2 min was followed by 30 cycles of 94°C for 10 s, 55°C for 15 s, and 72°C for 2 min. Finally the products were extended by incubation at 72°C for 10 min. MAM1 was amplified using the primers MAM1/49SacI and MAM1/TGAXhoI; MAML with MAML/49SacI and MAML/TGAXhoI; MAML-3 with MAML-3/46BglII and MAML-3/TGAPstI; and finally MAML-4 with MAML-4/57SacI and MAML-4/TGAXhoI. After amplification the MAML-3 product was digested with BglII and PstI, while the MAM1, MAML, and MAML-4 products were digested with SacI and XhoI and gel purified. The products were then ligated into the inducible expression vector pQE30 (Qiagen) digested with BamHI and PstI in the case of MAML-3 to give pQE-MAML-3, or SacI and SalI in the case of MAM1, MAML, and MAML-4 to give pQE-MAM1, pQE-MAML and pQE-MAML-4. The four constructs and the empty vector were used to transform E. coli strain m15 (Qiagen) containing the Lac repressor plasmid pREP4. The constructs were isolated and sequenced to ensure no mutations had been introduced.

E. coli CV512 Complementation

Plasmids pQE-MAM1, pQE-MAML, pQE-MAML-3, pQE-MAML-4, and an empty vector control were transferred into the Leu auxotrophic E. coli strain CV512 obtained from the CGSC E. coli Genetic Stock Centre (CGSC no. 5539) and containing pREP4. To test for complementation CV512 containing each of the constructs grown on Luria-Bertani medium was streaked onto plates of solid M9 media (1× M9 salts [48 mm Na₂HPO₄, 22 mm KH₂PO₄, 8.5 mm NaCl, and 18.7 mm NH₄Cl]; 2% [w/v] Glc; 1 mm thiamine; 1 mm MgSO₄; 0.1 mm CaCl₂; kanamycin, 50 μg mL⁻¹; and carbenicillin, 100 μg mL⁻¹), M9 media supplemented with 0.1 mm isopropyl-β-d-thiogalactopyranoside (IPTG), and M9 media supplemented with 0.3 mm Leu. The plates were incubated overnight at 37°C before being photographed. This was performed at least three times. CV512 containing each construct was also used to inoculate the same liquid M9 medium to an OD₆₀₀ of 0.01 and incubated at 37°C with shaking. Growth curves were obtained by measuring the OD₆₀₀ at the times indicated in Figure 2b. The growth curves were repeated three times and the average OD₆₀₀ used.

Protein Analysis

CV512 containing pQE-MAM1, MAML, MAML-3, MAML-4, or an empty vector control was used to inoculate 10 mL of Luria-Bertani medium and induced with 0.1 mm IPTG at mid-log phase. After 2 h total protein was extracted, separated by SDS-PAGE (Laemmli, 1970), immunoblottted with an anti-His horseradish peroxidase conjugated antibody (Invitrogen), and detected with the SuperSignalWest Pico chemiluminsecent system (PerBio, Tattenhall, UK) according to the manufacturer's instructions.

Isolation of MAM Insertion Lines

The MAML insertion line, MAML KO, was identified by PCR screening an En-1 mutagenized population of Arabidopsis, ecotype Col-0 (Baumann et al., 1998; Wisman et al., 1998a, 1998b) using forward and reverse primers specific to the MAML gene (MAML/F and MAML/R) and the En-1 transposable element (EN/F and EN/R). After back-crossing once to Col-0 a line was isolated that contained a single copy of the En-1 transposon, inserted into MAML. The MAML-4 T-DNA insertion line, Garlic1175_E02, was identified by screening the SAIL/GARLIC T-DNA insertion population of Arabidopsis ecotype Col-0 (Sessions et al., 2002). Homozygous individuals were generated for each line, and compared with the relevant wild type in all subsequent analyses. Col-0 and ecotype Wassilewskija seed was obtained from the Nottingham Arabidopsis Stock Centre (UK).

DNA Extraction and PCR Analysis of Insertion Sites

DNA was extracted from the leaves of each insertion line using a modified version of the cetyl-trimethyl-ammonium bromide method (Lister et al., 2000). Sequencing was performed using the ABI BigDye dye terminator system (Perkin-Elmer Applied Biosytems, Foster City, CA). The T-DNA insertion site for Garlic1175 was confirmed by PCR and sequencing using T-DNA primers (GARLICLB1 and GARLICRB1) and MAML-4 specific primers (MAML-4/R1 and MAML-4/L1). The MAML::En insertion site was further analyzed by PCR and sequencing using various MAML specific primers and En-1 specific primers (not listed).

Semiquantitative RT-PCR Analysis

For semiquantitative RT-PCR analysis, five microliters of cDNA dilution was used in a 50-μL PCR mixture containing 1 unit Taq polymerase (Invitrogen) and 0.15 μm each primer; the reaction was allowed to proceed for 18 cycles. The primer pairs used are shown below. The constitutively expressed housekeeping gene adenine phosphoribosyltransferase (APT) was used as an mRNA loading control (Moffatt et al., 1994). Each reaction was repeated at least three times independently. The products were then blotted onto a positively charged nylon membrane (Hybond-N+, Amersham Pharmacia, Uppsala) and probed with ³²P-random-labeled probes according to established procedures. Probes were derived from plasmids (pGEMTEasy, Promega, Southampton, UK) that contained the sequenced product from each primer pair used in the RT-PCR reaction.

GSL Analysis

GSLs were extracted from 300 mg of seeds, converted to desulphoglucosinolates, and analyzed by LC-MS with atmospheric pressure chemical ionization, as previously described (de Quiros et al., 2000).

MAML Cloning and Arabidopsis Transformation

Transgenic Arabidopsis were generated by Agrobacterium tumefaciens mediated transformation with the T-DNA vector pTKC28 through floral dipping (Clough and Bent, 1998) and selection of the seedlings for BASTA resistance. The Agrobacterium host was a rifampicin resistant derivative of C58 containing a nononcogenic Ti plasmid, supplying the virulence functions and an intermediate vector containing the genes of interest between the T-DNA borders. These consisted of the bialophos resistance gene (bar) from Streptomyces hygroscopius driven by the Arabidopsis small subunit Rubisco promoter as a marker gene, and the MAML gene consisting of a Sfc/HincII 3,379 bp genomic fragment derived from bacterial artificial chromosome clone T20O7 (Arabidopsis Biological Resource Center, Columbus, OH) containing the complete open reading frame (including introns) plus 339 bp downstream of the stop codon driven by the 35S CaMV promoter.

Soluble Amino Acid Analysis

A total of 100 mg of 9 d-old seedlings was homogenized in 1 mL of 45°C 70% methanol and incubated at 45°C for 10 min. The supernatant was removed and the pellet extracted twice more and the supernatants pooled. The supernatants were dried under a fixed nitrogen line at 45°C and resuspended in 0.02 m HCl. The extract was then filtered through an ultrafree-MC 0.22-μm filter column (Millipore, Bedford, MA). Ten microliters of eluate was derivatized with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate using the AccQ Tag system (Waters, Milford, MA). The derivatized amino acids were separated on a Waters Alliance 2695 Separation Module through a reverse phase AccQ Tag column (Waters) at 37°C using a 65-min gradient of sodium acetate buffer (0.1 m sodium acetate pH 5.80, 2.7 μm EDTA, and 6.9 mm triethylamine), acetonitrile, and water at a flow rate of 1 mL min⁻¹. Derivatized amino acids were detected by excitation at 250 nm and emission at 395 nm using a Waters 474 scanning fluorescence detector. Millenium³² Chromatography Manager software (Waters) was used to analyze the data. Individual amino acids were identified and quantified using a calibration curve generated by the injection of standards of known concentrations. Four extractions were performed for each line analyzed to ensure reproducibility.

For LC-MS/MS analysis of derivatized amino acids, derivatized extracts were separated on a Thermofinnigan Surveyor HPLC (Hemel Hempstead, UK) through a reverse phase Luna C18 column (Phenomonex, Cheshire, UK) using a gradient of solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in methanol) at a flow rate of 300 μL/min throughout. Two gradients were used. Gradient 1 was as follows: 30% solvent B at start, linear gradient to 70% solvent B over 26.6 min, linear gradient to 97% solvent B over 2.4 min, linear gradient to 30% solvent B over 0.2 min, and 30% solvent B for 6.8 min. Gradient 2 was as follows: 5 min at 7% solvent B, linear gradient to 40% solvent B over 20 min, linear gradient to solvent 97% B over a further 20 min, and 97% solvent B for 5 min. Between runs the column was reequilibrated with 7% solvent B for 10 min. Derivatized amino acids were detected by UV A₂₅₀.

For LC MS/MS analysis of underivatized amino acids, untreated extracts were separated on a Thermofinnigan Surveyor HPLC through a reverse phase Luna C18 column using a gradient of solvent A (0.1% heptafluorobutyric acid in water; Sigma, St. Louis) and solvent B (acetonitrile) at a flow rate of 300 μL min⁻¹ throughout. The gradient was as follows: 10% solvent B at start, linear gradient to 25% solvent B over 20 min, linear gradient to 40% solvent B over 2.5 min, linear gradient to 10% solvent B over 0.5 min. Between runs the column was reequilibrated with 10% solvent B for 10 min. After column separation, amino acids in the derivatized and underivatized extracts were detected by positive mode electrospray ionization in a Thermofinnigan LCQ DecaXP ion-trap mass spectrometer. The source conditions were 5.2 kV source voltage, 350°C capillary temperature, 50 units sheath gas, no auxiliary gas. Ions of interest were selected for fragmentation using an isolation width of 2 m/z and collision energy of 35% without wideband activation.

Primers

All primers were synthesized by Sigma Genosys (Cambridge, UK).

For cDNA cloning:

• MAM1/49SacI 5′-ACGAGCTCTGCTCCGCTGAGTCCAAAAAG-3′

• MAM1/TGAXhoI 5′-GACCTCGAGCCAAACTTATACAACAGCGGAAA-3′

• MAML/49SacI 5′-ACGAGCTCTGCTCTTCTGTGTCCAAAAATG-3′

• MAML/TGAXhoI 5′-GACCTCGAGCGTGTTTACACATTCGATGAAA-3′

• MAML-3/46BglII 5′-ACAGATCTCTTACCACCGCCGGAAAATTC-3′

• MAML-3/TGAPstI 5′-ACCTGCAGTTTCTTCAGGCAGGGACTTC-3′

• MAML-4/57SacI 5′-ACGAGCTCTGCTCAATCTCAGATCCTTCTC-3′

• MAML-4/TGAXhoI 5′-GACCTCGAGTTCAGGCAGCGACTCTGTT-3′

Insertion site analysis:

• GARLICLB1 5′-GCCTTTTCAGAAATGGATAAATAGCCTTGCTTCC-3′

• MAML/F 5′-TATGCGAAGAGGCCGAGGGTAATG-3′

• MAML/R 5′-CTTTACCATACCCTGCCGACACATACC-3′

• EN/F 5′-AGAAGCACGACGGCTGTAGAATAGGA-3′

• EN/R 5′-GAGCGTCGGTCCCCACACTTCTATAC-3′

For RT-PCR analysis:

• MAM1/L1 5′-ACCACTAGCTGTCGCTCCAT-3′

• MAM1/R1 5′-CTTGGCGATGGTTTTAATAGC-3′

• MAML/L1 5′-CCGGTCAGTGTTACCCTTTC-3′

• MAML/R1 5′-TTGGCGATGGTCTTAATGGT-3′

• MAML-3/L1 5′-CTTATCCTCCTCCTCCACCT-3′

• MAML-3/R1 5′-CAGGGACATAGCCATTTTCG-3′

• MAML-4/L1 5′-TTTCCGTTTCCAACCATCTC-3′

• MAML-4/R1 5′-TCCTAGCGATTTCGATGACC-3′

• APT/F 5′-TCCCAGAATCGCTAAGATTGCC-3′

• APT/R 5′-CCTTTCCCTTAAGCTCTG-3′

Upon request, all novel materials described in this publication will be made available in a timely manner for noncommercial research purposes, subject to the requisite permission from any third-party owners of all or parts of the material. Obtaining any permissions will be the responsibility of the requestor.