Two chitinase-like proteins abundantly accumulated in latex of mulberry show insecticidal activity
© Kitajima et al. 2010
Received: 28 July 2009
Accepted: 28 January 2010
Published: 28 January 2010
Skip to main content
© Kitajima et al. 2010
Received: 28 July 2009
Accepted: 28 January 2010
Published: 28 January 2010
Plant latex is the cytoplasm of highly specialized cells known as laticifers, and is thought to have a critical role in defense against herbivorous insects. Proteins abundantly accumulated in latex might therefore be involved in the defense system.
We purified latex abundant protein a and b (LA-a and LA-b) from mulberry (Morus sp.) and analyzed their properties. LA-a and LA-b have molecular masses of approximately 50 and 46 kDa, respectively, and are abundant in the soluble fraction of latex. Western blotting analysis suggested that they share sequence similarity with each other. The sequences of LA-a and LA-b, as determined by Edman degradation, showed chitin-binding domains of plant chitinases at the N termini. These proteins showed small but significant chitinase and chitosanase activities. Lectin RCA120 indicated that, unlike common plant chitinases, LA-a and LA-b are glycosylated. LA-a and LA-b showed insecticidal activities when fed to larvae of the model insect Drosophila melanogaster.
Our results suggest that the two LA proteins have a crucial role in defense against herbivorous insects, possibly by hydrolyzing their chitin.
Laticifer cells of plants are unique in shape, development and function. These cells form long tubular or branched structures that run through the plant body, and large amounts of cytoplasm are exuded when the plant body is cut. The cytoplasm is a sticky white fluid called latex, and in the case of the para rubber tree (Hevea brasiliensis) and rubber fig (Ficus elastica), insoluble latex particles are a well-known commercial source of natural rubber . Laticifers are not present in all plant species. For example, well-studied model plants such as Arabidopsis, rice, wheat, and tobacco, do not have laticifers. However, they have been reported in about 12 500 plant species in 22 families, including monocots and dicots, and are estimated to exist in up to 20 000 species in 40 families .
One important feature of laticifers is that they contain various toxic compounds in the latex; for example, the neurotransmitter dopamine in the Persian poppy (Papaver bracteatum), narcotic alkaloid morphine in the opium poppy (Papaver somniferum), and insecticidal compounds such as the glycosidase inhibitors 1,4-dideoxy-1,4-imino-d-arabinitol (d-AB1) and 1-deoxynojirimycin (DNJ) in mulberry [2, 3]. In addition, cysteine protease in latex of papaya (Carica papaya) and wild fig (Ficus virgatalatex) is toxic to caterpillars of herbivorous insects . The compounds are often highly condensed in the latex; for example, d-AB1, DNJ and its related compound 1,4-dideoxy-1,4-imino-D-ribitol are present at 8-18% dry weight in latex of mulberry leaf; 100 times higher than that in the whole leaf . Rubber is also toxic to herbivorous insects because its stickiness limits movement of their mouthparts .
These findings suggest that laticifers are specialized cells that have a role in defending plants against herbivorous insects. The distribution of laticifers is associated with vascular tissue, which transports metabolites, water, minerals and signaling molecules through the plant body, and is one of the most important plant tissues. Laticifers might protect the vascular system. Long tubular or branched network structures are effective in this role, because they enable laticifers to exude large amounts of latex at damaged sites. In the present study, we speculate that proteins that are abundant in latex have a defensive role against herbivorous insects. Here, we report that two proteins that are abundant in the soluble fraction of mulberry latex have insecticidal activities against the model insect Drosophila melanogaster, possibly through their chitinase activity.
LA-a and LA-b also were able to hydrolyze chitosan, which is partly deacetylated and therefore solubilized chitin. Chitosanase activity was tested at a pH range of 3-7, and maximum activity was observed at pH 5 (Figure 5B). At pH 5, the chitosanase activity of LA-a and LA-b was 1.8 and 1.4 times higher, respectively, than that of hen egg lysozyme (data not shown). In the report by Mayer et al. , chitosanase activity of 2 chitinase isoforms was 30 and 100 times higher than that of hen egg lysozyme, and the activity of 4 isoforms was 1.7-3.5 times higher. One isoform showed no chitosanase activity.
These results indicated that LA-a and LA-b had chitinase and chitosanase activities, although the activities were relatively low compared to those of orange chitinases.
In this study, we showed that the two proteins LA-a and LA-b were abundant in the soluble fraction of mulberry latex, and had low, but significant chitinase and chitosanase activities. In addition, both LA proteins showed insecticidal activity against larvae of D. melanogaster. The finding that these major proteins have insecticidal activity supports the idea that laticifers are specialized cells for defense against herbivorous insects.
In a recently published study  on mulberry latex protein MLX56, it has been shown that MLX56 is glycosylated and has growth-inhibiting (not insecticidal) activity on the lepidopteran caterpillars of cabbage armyworm (Mamestra brassicae) and Eri silkworm (Samia richini), but not on mulberry specialist silkworm (Bombyx mori). Similarities between LA-a and LA-b proteins and MLX56, with respect to N-terminal amino acid sequence and electrophoretic mobility in SDS-PAGE, suggest that they are the same or similar proteins. However, contrary to our results, MLX56 has been reported to have no chitinase activity . Enzyme activity of MLX56 may need to be analyzed using various substrates. In our study, we observed chitinase activity of LA-a and LA-b when using CM-chitin-RBV and chitosanase activity when using chitosan (Figure 5), but the activity was hardly detectable when 4-methylumberryferyl β-D-N, N', N"-triacetylchitotriose (4-MU-(GlcNAc)3), a common substrate for plant chitinase assay, was used (data not shown). Substrate specificity of LA proteins and MLX56 could be different from that of common plant chitinases.
Chitin, a substrate of chitinase, forms the cuticle of the exoskeleton in insects, and should be the target of LA-a and LA-b in the larvae of D. melanogaster and other insects. In addition to external body surfaces, the inside of the mouthpart is also a possible target of LA-a and LA-b because the pH inside the mouthpart is expected to be near that of crude extract of mulberry tissues (weekly acidic or neutral) and to be optimal for the chitinase activity of LA proteins.
Plant chitinase is well known to have antifungal activity, but only a few studies have investigated its toxicity to insects. It has been reported that transgenic tomato that accumulated poplar (Populas sp.) chitinase inhibited development of Colorado potato beetle (Leptinotarsa decemlineata) larvae, and the amount of chitinase in the leaves was >0.3% (w/w) . Two papers have reported that insect chitinase showed toxicity to insects; chitinase of tobacco hornworm (Manduca sexta) included at 2% in an artificial diet showed 100% mortality of larvae of merchant grain beetles (Oryzaephilis mercator) , and transgenic tobacco leaves that accumulated chitinase of tobacco hornworm at 0.02-0.03% of total leaf protein inhibited growth of tobacco budworm (Heliothis virescens) larvae .
The insecticidal activity of LA proteins might not be caused simply by chitinase activity. It could be that a domain of LA proteins with unidentified biochemical activity cooperates with the chitinase domain to enhance insecticidal activity. Estimated molecular masses of LA-a and LA-b (50 and 46 kDa) were higher than those of common plant chitinases (20-40 kDa). This is at least partly caused by glycosylation, but it also suggests that the proteins might have another functional domain and biochemical function, working in cooperation with chitinase activity for efficient toxicity. Further analysis is needed to understand the molecular mechanism of insecticidal activity of LA-a and LA-b.
Unlike some other insects that eat leaves of latex-producing plants, silkworm eats mulberry leaves without removing the latex by cutting or trenching veins (see [13, 14] for reviews). Therefore, silkworms probably have a mechanism for detoxifying the insecticidal activity of LA proteins, and are unsuitable for studying potential insecticidal activity of LA proteins. One of the reasons why we used D. melanogaster in this study is that this insect does not eat mulberry and thus is not expected to overcome potential toxicity of LA proteins. It would be interesting to know to which species of insect other than D. melanogaster (e.g. pest insects of economically important corps) the LA proteins are toxic. Some insect species including silkworm would be tolerant to LA proteins, and it would be interesting to investigate the mechanism by which they can overcome the toxicity of the LA proteins.
In the present study, we indicated that two chitinase-like proteins (LA-a and LA-b) were abundantly accumulated in latex of mulberry and suggested that they have a crucial role in defense against herbivorous insects, possibly by hydrolyzing their chitin.
Mulberry (Morus alba L. cv. Minamisakari) was maintained at the Center for Bioresource Field Science, Kyoto Institute of Technology.
Latex exuded from the cut petiole of mulberry was mixed rapidly with an equal volume of buffer A (100 mM potassium phosphate, 10 mM EDTA, pH 6.7) supplemented with 0.1% (v/v) β-mercaptoethanol, frozen in liquid nitrogen, and stored at -80°C. Six milliliters of the latex/buffer A mixture was diluted 4 times with buffer A and centrifuged at 18 000 g for 30 min. The supernatant was fractionated by ammonium sulfate between 30 and 80% saturation. Precipitated protein was dissolved in 4 ml of buffer B (10 mM potassium phosphate, 1 mM EDTA, pH 6.0) and dialyzed against the same buffer overnight, using a dialysis membrane with a molecular weight cut-off of 25 kDa (Spectra/Por 7; Spectrum Laboratories Inc., Rancho Dominguez, CA, USA). The sample was diluted to 50 ml and loaded onto a CM-Cellulofine C-200 (Seikagaku Kogyo KK, Tokyo, Japan) cation-exchange column (26 mm i.d. × 100 mm) equilibrated with buffer B. Protein fractions that contained LA-a and LA-b were eluted with buffer B that contained 0.2 M KCl at a flow rate of 1.0 ml min-1. Eluted protein solution was brought to 30% saturation with ammonium sulfate, and hydrophobic interaction chromatography was carried out on a HiLoad 16/10 phenyl sepharose HP column (16 mm i.d. × 100 mm; GE healthcare, Piscataway, NJ, USA) equilibrated with buffer B that contained ammonium sulfate at 30% saturation. The column was developed with a linear gradient of 30-0% saturation of ammonium sulfate at a flow rate of 2 ml min-1. Fractions that contained LA-b were desalted twice by gel filtration chromatography using PD10 (GE healthcare) and 10DG (Bio-Rad, Hercules, CA, USA) column, and further purified on a UNOS1 cation-exchange column (Bio-Rad) eluted with a linear gradient of 0-0.5 M NaCl in buffer B at a flow rate of 3.0 ml min-1. Purified LA-a and LA-b proteins were concentrated with Microcon YM-10 (Millipore Corp., Bedford, MA, USA) to 1.0 mg ml-1, frozen with liquid nitrogen, and stored at -80°C. All purification steps were performed at 4°C. All chromatography steps were carried out using a BioLogic Duo-Flow chromatography system (Bio-Rad). Protein concentrations were determined according to the equation: protein concentration (mg ml-1) = 1.45 A280 - 0.74 A260 .
A rat polyclonal antibody was raised against purified LA-a by a standard procedure. For western blotting, protein samples were separated by SDS-PAGE and electroblotted onto polyvinylidene difluoride (PVDF) membrane using the buffer system of Kyhse-Anderson . The membrane was reacted with the polyclonal anti-LA-a antibody, followed by alkaline phosphatase-conjugated anti-rat antibody according to Coligan et al. . A mixture of 5-bromo-4-chloro-3-indolyl phosphate (BCIP) and nitro blue tetrazolium (NBT) (Promega, WI, USA) was used for color development.
Cysteine residues of proteins were alkylated by treatment with iodoacetamide according to Coligan et al. . Proteins were then purified on a C18 reversed-phase column (4.6 × 150 mm; TSK gel ODS-100S; Toso, Tokyo, Japan) eluted with a linear gradient of 5-50% acetonitrile in 0.1% trifluoroacetic acid at a flow rate of 1.0 ml min-1. Eluted proteins were concentrated by evaporation and analyzed by automated protein sequencer PPSQ-21A (Shimadzu, Japan).
Protein samples were separated by SDS-PAGE and the gel was stained, first with Pro-Q Emerald 300 to detect glycosylated proteins, and then with SYPRO Ruby stain (both from Invitrogen, Carlsbad, CA, USA) for total proteins, according to the manufacturer's instructions. As a control, CandyCane glycoprotein molecular weight standards (Invitrogen) were used. For the lectin blot, protein samples were separated by SDS-PAGE and electroblotted onto a PVDF membrane as described above. The membrane was reacted with biotinylated RCA120 (Vector Laboratories, Burlingame, CA, USA), and subsequently with alkaline phosphatase-conjugated streptavidin. A mixture of BCIP and NBT was used for color development. Fifteen other lectins (ConA, DBA, LCA, PHA-E4, PNA, RCA120, UEQ-I, WGA, ABA, DSA, Lotus, MAM, PHA-L4, SBA, and SSA) were purchased from Seikagaku Kogyo (Tokyo, Japan).
Colorimetric chitinase assay was performed using CM-chitin-RBV (Loewe Biochemica, München, Germany) as a substrate, according to Mayer et al. , except that the enzyme reaction was prolonged from 10 min to 2 h. Buffers used were 0.1 M citrate-Na2HPO4 buffers (pH 5-8). Hydrolyzed chitin was recovered and measured by absorbance at 550 nm. Activity was expressed as Δ550 nm g protein-1 min-1. Fluorometric chitosanase assay was performed using soluble chitosan (Chitosan, low molecular weight; Sigma-Aldrich, MO, USA) as a substrate according to Osswald et al. , except that the enzyme reaction was prolonged from 30 min to 2 h. Buffers used were 0.1 M citrate-Na2HPO4 buffers (pH 3-7). Controls consisted of the complete reaction mixtures stopped immediatedly after addition of enzyme, as described by Osswald et al. . The reaction product glucosamine (GlcN) was recovered, labeled with fluorescamine (Fluka, Buchs, Switzerland) and monitored fluorometrically (excitation: 395 nm; emission: 493 nm). Chitosanase activity was calculated on the basis of GlcN equivalents using a GlcN standard curve generated under the same conditions as these used for the assays.
Drosophila melanogaster (Canton-S strain) was reared at 25°C on standard yeast agar medium. To test insecticidal activity of LA-a and LA-b proteins, 15 2-day-old larvae (second instar) were placed in a vial that contained 200 mg Formula 4-24 instant Drosophila medium (Carolina Biological Supply, Burlington NC, USA), rehydrated with 0.8 ml PBS supplemented with 1 mg LA protein and 10 mg Bacto Yeast Extract (Difco, Detroit, MI, USA) as a nitrogen source. This accounted for 0.1% (w/w) LA protein. For control experiments, the diet was supplemented with the same amount of bovine serum albumin (BSA) or no protein instead of LA protein. The vials were incubated at 25°C.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.