Extracellular Expression of a Functional Recombinant Ganoderma ...

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 2008, p. 1039–1049 0099-2240/08/$08.00⫹0 doi:10.1128/AEM.01547-07 Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Vol. 74, No. 4

Extracellular Expression of a Functional Recombinant Ganoderma lucidium Immunomodulatory Protein by Bacillus subtilis and Lactococcus lactis䌤† Chuan M. Yeh,* Chun K. Yeh, Xun Y. Hsu, Qiu M. Luo, and Ming Y. Lin Department of Food Science and Biotechnology, National Chung-Hsing University, Taichung, Taiwan, Republic of China Received 9 July 2007/Accepted 11 December 2007

Bacillus subtilis and Lactococcus lactis are ideal hosts for the production of extracellular heterologous proteins of major commercial importance. A recombinant gene for the novel Ganoderma lucidium immunomodulatory protein LZ-8, recombinant LZ-8, was designed encoding the same amino acid sequence but using the preferred codons for both strains and was synthesized by overlapping extension PCR. Using the signal peptide (SP) from subtilisin YaB (SPYaB), recombinant LZ-8 was expressed extracellularly in Bacillus subtilis and Lactococcus lactis. In the absence of SPYaB, recombinant LZ-8 was expressed extracellularly in B. subtilis, but not in L. lactis. The three expressed recombinant LZ-8s showed different capacities for modulating the production of Th1 and Th2 cytokines by peripheral blood mononuclear cells and of tumor necrosis factor alpha by a macrophage cell line.

protein production, extracellular production is a convenient way to avoid contamination from cellular proteins and makes possible continuous cultures for downstream processing (10). Gram-positive strains, such as Bacillus subtilis (55) and Lactococcus lactis (6), are widely used as hosts for the extracellular production of heterologous proteins. Both strains are able to secrete proteins, and genetic tools for each host are available. B. subtilis is a generally recognized as safe (GRAS) bacterium and has been widely used in the production of heterologous proteins (16, 17, 53, 58). L. lactis has long been widely used in food production (27) and can be used as a food-grade host in heterologous protein production. A variety of new applications of L. lactis have been developed, such as the expression of cytokines (1), antigens (38), enzymes (2), and membrane proteins (25) and the metabolic engineering of L. lactis (11). A food-grade L. lactis expression system has been developed for industry-scale production (32). To produce recombinant LZ-8 for oral consumption, GRAS or food-grade hosts and extracellular expression are recommended. In this study, a novel gene coding for recombinant LZ-8, recombinant LZ-8, was designed using the preferred codons for both B. subtilis and L. lactis and synthesized by overlap extension PCR (OE-PCR). To allow the secretion of recombinant LZ-8 from B. subtilis or L. lactis cells into the extracellular medium, the codon-optimized recombinant LZ-8 gene was fused to optimal promoters fused to a DNA fragment encoding the signal peptide (SP) from subtilisin YaB (SPYaB). The growth media were optimized to obtain higher expression levels. Recombinant LZ-8s were purified from the extracellular media and used in cytokine stimulation assays to examine the immunomodulatory functions of the proteins from various recombinant sources.

Ling Zhi (Ganoderma lucidium, a species of Basidiomycetes) has a long history of use in China as a traditional medicine for the treatment of various ailments. A novel protein named Ling Zhi-8 (LZ-8), a member of the fungal immunomodulatory proteins (FIPs), has been purified from the mycelial extract of Ling Zhi (19), and the biological activities of native LZ-8 have been studied extensively. In terms of antitumor activity, LZ-8 acts as a mitogen for mouse spleen cells (35), human peripheral lymphocytes (9), and human peripheral blood mononuclear cells (hPBMCs) (50), and a polysaccharide fraction containing 5% peptide from Ganoderma lucidium induces the production of cytokines and tumor necrosis factor (TNF) by activated macrophages and T lymphocytes (52). In terms of immunomodulatory responses, LZ-8 modifies the expression of adhesion molecules on immunocompetent cells (33) and suppresses autoimmune diabetes in young female nonobese diabetic mice (20). In a transplanted allogeneic pancreatic rat islet model, LZ-8 delays the rejection of the allografted islets (49). Dimerization of the protein, mediated by the N-terminal amphipathic ␣-helix, is important in the immunomodulatory activities of FIPs (30, 39). LZ-8 is therefore considered a good candidate for development as a functional food supplement or therapeutic agent for use in the prevention or treatment of cancer and autoimmune diseases. Currently, fermentation of Ganoderma lucidium mycelia and purification of LZ-8 are the main method used to obtain LZ-8, but this process is costly and time consuming and results in a low yield. High-level production of recombinant LZ-8 would be a good way to overcome these problems. In recombinant

* Corresponding author. Mailing address: Department of Food Science and Biotechnology, National Chung-Hsing University, Taichung, Taiwan, Republic of China. Phone: 886-4-22840386, ext. 3190. Fax: 886-4-22853165. E-mail: [email protected]. † Supplemental material for this article may be found at http://aem .asm.org/. 䌤 Published ahead of print on 21 December 2007.

MATERIALS AND METHODS Bacterial strains, plasmids, and culture medium. The bacterial strains and plasmids used are shown in Table S1 in the supplemental material. The Escherichia coli DNA manipulation host strains were propagated at 37°C with agitation

1039

1040

YEH ET AL.

APPL. ENVIRON. MICROBIOL.

FIG. 1. OE-PCR schematic (A) and OE-PCR product (B). In panel A, the solid lines indicate primers and the dashed lines the PCR amplification fragments. A BamHI site and an SalI site were introduced into the synthesized recombinant LZ-8 gene. In panel B, lane M contains the 100-bp marker and lane 1 the OE-PCR products. The arrowhead indicates the predicted PCR product.

in Luria-Bertani (LB) broth (Difco), supplemented with 25 ␮g/ml of tetracycline for transformants with B. subtilis/E. coli shuttle vector pHY300PLKderived plasmids or with 20 ␮g/ml of chloramphenicol for transformants with L. lactis/E. coli shuttle vector pNZ8008-derived plasmids. B. subtilis DB104 (18) or B. subtilis WB800 (4) cells were used as the B. subtilis recombinant LZ-8 expression host. The B. subtilis transformants were cultured in LB medium or medium A (36) containing 10 ␮g/ml of tetracycline. L. lactis NZ9000 (21) or L. lactis NZ9000 ⌬htrA (34) was used as the L. lactis recombinant LZ-8 expression host and propagated at 30°C in M17 broth (Difco). L. lactis transformants harboring pNZ8008-derived plasmids were cultured with 5 ␮g/ml of chloramphenicol in GM17 medium (M17 broth supplemented with 0.5% glucose), GM17B medium (GM17 medium buffered at pH 6.9 with 0.2 M potassium phosphate), 2GM17B medium (M17 broth supplemented with 1% glucose and buffered at pH 6.9 with 0.2 M potassium phosphate), or 2NGM17B medium (GM17B medium containing double the concentration of peptones and yeast extract). Cell growth was monitored by measuring the cell density at an optical density of 600 nm. Molecular techniques. Plasmid DNA was isolated by using a Mini-M plasmid DNA extraction system (Viogene, Taipei, Taiwan). Lactic acid bacterium genomic DNA was isolated by using a blood and tissue genomic DNA extraction system (Viogene). PCR products were purified by using a PCR-M cleanup system (Viogene). DNA fragments were recovered from gels by using a Gel-M gel extraction system (Viogene). Plasmids, genomic DNA, and PCR products were analyzed by agarose gel electrophoresis (42). The primers were purchased from Invitrogen Life Technologies (Carlsbad, CA). Ex Taq DNA polymerase (Takara, Bio Inc., Kyoto, Japan) was used for PCR amplifications. Electrotransformation of E. coli was carried out at a field strength of 12.5 kV/cm, capacitance of 25 ␮F, and resistance of 200 ⍀ using a Gene Pulser II electroporation apparatus (Bio-Rad, Hercules, CA). The electrotransformation of B. subtilis (56) and L. lactis was performed as described previously (57). Design and synthesis of the novel recombinant LZ-8 gene. The novel recombinant LZ-8 gene was designed using the most-preferred codons for both B. subtilis and L. lactis (codon usage database, GenBank-NCBI; http://www.kazusa .or.jp/codon/) and was synthesized by OE-PCR. An extra BamHI sequence was added to the 5⬘ region, and six extra histidine codons, a stop codon, and an SalI sequence were added to the 3⬘ region (Fig. 1). The details of the design and the primers used can be found in a pending patent (C. M. Yeh, 4 April 2006, Republic of China patent pending. Application 095112001.). The LZ-BamHI, LZ-1, LZ-2, LZ-3, LZ-4, LZ-5, LZ-6, and LZ-SalI primers were designed with each primer being partially overlapped by complementary sequences (Fig. 1). The OE-PCR reaction for amplifying recombinant LZ-8 was performed by using Ex Taq DNA polymerase (Takara, Kyoto, Japan) and conditions of 94°C for 2

min and 35 cycles of 94°C for 30 s, 60°C for 30 s, and 72°C for 2 min, followed by 72°C for 5 min. Plasmid construction. The construction of recombinant LZ-8 expression vectors was preceded by serial subcloning of the PCR-amplified recombinant LZ-8 gene. The PCR-amplified recombinant LZ-8 gene was digested with BamHI and SalI and inserted into plasmid p3ATT (an E. coli plasmid pET29a derivative containing the T7 promoter [PT7]–antifreeze protein gene [afp3]–six-histidinecodon [his6]–terminator [terQ]) or plasmid pSP3ATT (pET29a derivative, PT7– SPYaB–afp3–his6–terQ, in which an antifreeze protein gene was fused after SPYaB) digested with the same enzymes, resulting in pLZ or pSLZ, respectively. The recombinant LZ-8 gene was fused to the SPYaB gene (SPYaB–recombinant LZ-8) in pSLZ. Plasmid pLZ or pSLZ was then digested with NdeI and XbaI, and the recombinant LZ-8 or SPYaB–recombinant LZ-8 fragment inserted into pSECS-4ALET (51) digested with the same enzymes, resulting in pHLZ or pHSLZ, in which the recombinant LZ-8 gene or SPYaB–recombinant LZ-8 fused gene can be expressed from a synthetic expression control sequence 4 (SECS-4) and terminated at a DNA fragment containing the lambda t0 and rrnB T1 terminators in B. subtilis (51). Two other expression vectors were constructed. The recombinant LZ-8 or SPYaB–recombinant LZ-8 fragment was restriction digested from pHLZ or pHSLZ using NdeI and XbaI and ligated with pOAGFP⫹ (E. coli and B. subtilis shuttle vector pHY300PLK derivative containing SECS-2–green fluorescent protein [GFP] gene [gfp⫹]–terQ) treated with the same enzymes, resulting in pOA-LZ8 and pOAS-LZ8 in which recombinant LZ-8 or recombinant LZ-8 fused to the DNA fragment encoding the SPYaB sequence was expressed from the less-efficient SECS-2 and terminated at a DNA fragment containing the lambda t0 and rrnB T1 terminators in B. subtilis (51). pHLZ and pOA-LZ8 were expected to express recombinant LZ-8 in the cytosol fraction and pHSLZ and pOAS-LZ8 to express recombinant LZ-8 in the extracellular fraction. L. lactis expression vectors were then constructed using the above B. subtilis vectors. pHLZ or pHSLZ was digested with BglII and XbaI and the fragment SECS-4–recombinant LZ-8 or SECS-4–SPYaB–recombinant LZ-8 inserted into pNZter (31) digested with the same enzymes, resulting in pZLZ and pZSLZ. pZLZ and pZSLZ are L. lactis expression vectors in which recombinant LZ-8 was expressed from SECS-4 and which were expected to express recombinant LZ-8 at high levels in the cytosol and extracellular fractions, respectively. Two other, more-efficient L. lactis recombinant LZ-8 expression vectors, pNZLZ and pNZSLZ, were constructed by serial subcloning. The recombinant LZ-8–ter or SPYaB–recombinant LZ-8–ter fragments of pOA-LZ8 and pOAS-LZ8 were obtained by NdeI and XbaI digestion and ligated with pHYSA1-1ATT treated with the same enzymes, resulting in pHYLZ and pHYSLZ, in which the recombinant

VOL. 74, 2008

EXTRACELLULAR EXPRESSION OF BIOACTIVE RECOMBINANT LZ-8

LZ-8–ter or SPYaB–recombinant LZ-8–ter fragment was expressed from PslpA1, an efficient promoter in L. lactis (3, 37). pHYLZ or pHYSLZ was digested with XbaI and BglII, and the PslpA1–recombinant LZ-8–ter or PslpA1–SPYaB–recombinant LZ-8–ter fragments were ligated with pNZter treated with the same enzymes. The resulting plasmids, pNZLZ and pNZSLZ, expressed recombinant LZ-8 from the efficient PslpA1 promoter, and pNZSLZ secreted recombinant LZ-8 with the aid of SPYaB. pNZLZ and pNZSLZ were expected to efficiently express recombinant LZ-8 in the cell cytosol and extracellular fraction, respectively. Expression, purification, Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis, Western blot analysis, N-terminal amino acid analysis, and homodimer and circular dichroism (CD) analysis of recombinant LZ-8. Single colonies of B. subtilis or L. lactis transformed with various recombinant LZ-8 expression vectors were cultured, and the cell density measured at various time intervals. The cultures were centrifuged at 14,000 ⫻ g for 10 min at 4°C to remove the cell pellets. The supernatant proteins of B. subtilis cells were analyzed by Tricine-SDS-PAGE (43). The supernatant proteins of L. lactis were precipitated by using trichloroacetic acid (100 mg/ml final concentration), washed with cold acetone, air dried, and dissolved in sample buffer to a 10-fold concentration and then subjected to Tricine-SDS-PAGE analysis. An aliquot (16 ␮l) of supernatant was mixed with 4 ␮l of loading buffer, and the mixture was boiled for 5 min and then loaded in each lane of an SDS-PAGE gel. For recombinant LZ-8 purification, the supernatant proteins of the transformants were precipitated by 90% ammonium sulfate, dialyzed against deionized water, and purified by Ni-nitrilotriacetic acid (NTA) affinity chromatography in a native elution buffer according to the QIAexpressionist manual (4th ed.; Qiagen); the imidazole in the elution buffer was then removed by dialysis. The purified recombinant LZ-8 was quantified by using a DC protein assay kit (Bio-Rad Laboratories, Inc., United States). The levels of production of extracellular recombinant LZ-8 were estimated by comparing the densities of the recombinant LZ-8 bands in SDS-PAGE gels analyzed using Gel-Pro analyzer version 3.0 (Total-Integra Technology Co., Ltd., Taipei, Taiwan). For Western blot analysis, proteins in the SDS-polyacrylamide gels were transferred to a polyvinylidene fluoride Immobilon-P membrane (Millipore, Bedford, MA) by using an electroblotter (model VEP-2; Owl Scientific, Inc., NH) and immunodetected with polyclonal rat antibody against recombinant LZ-8 as described previously (48) or with His-tagged monoclonal antibody (Clontech, Takara Bio, Inc., Shiga, Japan) according to the instructions provided with Amersham ECL Plex Western blotting combination packs (Amersham Biosciences, United Kingdom). N-terminal amino acid analysis was performed by electroblotting of the SDS-PAGE-separated proteins onto polyvinylidene difluoride membranes (Bio-Rad Laboratories, Ltd., United States) followed by automatic Edman degradation using a Procise protein sequencing system (Max Planck Institute for Molecular Genetics, Berlin, Germany). For homodimer analysis, recombinant LZ-8 was cross-linked by incubation for 30 min at room temperature with the cross-linking reagent BS3 (bis-[sulfosuccinimidyl]; Pierce Co., United States) using a 1:50 molar concentration ratio, and samples of the reaction mixtures were analyzed by Tricine-SDS-PAGE (30). CD analysis was preceded by using a CD spectropolarimeter (J-815; Jasco Co., Great Dunmow, United Kingdom). The spectra were recorded using a protein concentration of 1.0 mg/ml and 10-mm path lengths, and the results of each 6th scan from 190 to 250 nm were averaged. The results were recorded in millidegrees and converted to molars in M⫺1 cm⫺1. Mass spectrophotometry was preceded by matrix-assisted laser desorption ionization–time of flight mass spectrometry (ABI Voyager-DE Pro; Blue Lion Biotech). Preparation of human mononuclear cells and mouse macrophage cell cultures, effects of recombinant LZ-8s on cell proliferation, and detection of recombinant LZ-8-induced cytokines. Mononuclear cells were isolated from the peripheral blood of three healthy young volunteers by Ficoll-Hypaque gradient centrifugation, and the effects of recombinant LZ-8s on cell proliferation and cytokine production were determined as described previously (29). To determine the effects of recombinant LZ-8s on cell proliferation, the hPBMCs were adjusted to 1.6 ⫻ 106 cells/ml in TCM medium (Gibco, Inc., United States), then plated (190 ␮l/well) in 96-well microtiter plates with different recombinant LZ-8 preparations at 0- to 50-␮g/ml final concentration or phosphate-buffered saline (PBS) buffer (137 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.5 mMKH2PO4, pH 7.2 to 7.4, filtered at 0.2-␮m pore size) as control, and incubated for 72 h at 37°C in a humidified incubator with 5% CO2/95% air. Twenty microliters of 5 mg/ml 3-(4,5-dimethylthiazole-2-yl)-2,5diphenyltetrazolium bromide (MTT) (catalogue no. M-5655; Sigma-Aldrich Co.) in PBS was then added to each well, the plates incubated at 37°C in a humidified incubator with 5% CO2/95% air for another 4 h and centrifuged at 200 ⫻ g for 10 min, and the culture medium discarded. One hundred microliters of dimethyl

1041

sulfoxide was added to each well, the plate was shaken on an oscillator for 30 min, and then the absorbance was measured at 550 nm on a plate reader (enzyme-linked immunosorbent assay [ELISA] reader; ASYS Hitech GmbH, Austria). For cytokine detection, the hPBMCs were washed three times with RPMI 1640 (Gibco, Germany) and resuspended in the same medium, adjusted to 1.6 ⫻ 106 cells/ml. An aliquot (950 ␮l) was placed in each well of a 24-well plate and incubated for 48 h with 50 ␮l of various concentrations of recombinant LZ-8 or PBS buffer in 5% CO2 at 37°C. The interleukin-2 (IL-2) in the culture supernatants was quantified by using a mouse IL-2 ELISA set (BD Biosciences Pharmingen, San Diego, CA), the sensitivity for IL-2 being ⬍3.1 pg/ml, while IL-4 was measured by using a mouse DuoSet ELISA development system (R&D Systems, Inc., Mineapolis, MN) with a sensitivity of about 15.6 pg/ml. The absorbance at 450 nm was then measured on a plate reader (ELISA reader; ASYS Hitech GmbH, Austria) and compared to those for a seven-point standard. The murine macrophage cell line RAW 264.7 (Riken Cell Bank, Tsukuba, Japan) was maintained in RPMI 1640 medium containing 5% heat-inactivated fetal calf serum (Gibco-BRL, Gaithersburg, MD) and antibiotics at 37°C under 5% CO2. RAW 264.7 cells were cultured in 96-well plates at a density of 1 ⫻ 105 cells and treated without or with various concentrations of recombinant LZ-8. The tumor necrosis factor alpha (TNF-␣) levels in the culture supernatants were measured by using a mouse DuoSet ELISA development system (R&D Systems, Inc., Minneapolis, MN) following the manufacturer’s protocol. Statistical analysis. The data are expressed as the means ⫾ standard deviations, and differences among samples were tested for significance using Duncan’s multiple range test and the Statistical Analysis System (SAS, release 8.2) program (SAS Institute, Inc., Cary, NC).

RESULTS Extracellular expression of recombinant LZ-8 with or without the aid of SP in Bacillus subtilis. The recombinant LZ-8 genes on B. subtilis expression vectors pHLZ and pHSLZ were transcribed from SECS-4, which was proved extremely efficient in B. subtilis, and terminated at a DNA fragment containing the lambda t0 and rrnB T1 terminators (51). pHLZ and pHSLZ were electrotransformed into B. subtilis DB104, a host deficient in three extracellular proteases. The transformants B. subtilis DB104(pHLZ) and B. subtilis DB104(pHSLZ) were expected to express recombinant LZ-8 at relatively high levels in the cell cytosol and extracellular fractions, respectively. The expected increase bands were not apparent on SDS-PAGE gels, but were seen on Western blots (Fig. 2). In the B. subtilis DB104(pHSLZ) transformant, the SP was not cleaved in the cell lysate, and recombinant LZ-8 was successfully processed and secreted into the medium (Fig. 2A). Surprisingly, the transformant B. subtilis DB104(pHLZ) expressed recombinant LZ-8 extracellularly without the aid of SPYaB (Fig. 2B). Improved expression of recombinant LZ-8 in B. subtilis. To improve the levels of expression of recombinant LZ-8, vectors pOA-LZ8 and pOAS-LZ8, which were transcribed from the less-efficient SECS-2 (51), were electrotransformed into B. subtilis WB800, an eight-protease-deficient host. The expression levels for both were higher than those in B. subtilis DB104. On SDS-PAGE gels, increased bands of recombinant LZ-8 of the expected size were seen in the medium after 12 h of culture of pOA-LZ8 or pOAS-LZ8 plasmid-harboring transformants, and these results were confirmed by Western blot analysis (data not shown). The expression levels were improved, and recombinant LZ-8 was expressed extracellularly with or without the SP, as seen with B. subtilis DB104(pHLZ) and B. subtilis DB104(pHSLZ). To enhance the production of recombinant LZ-8, various media were tested. B. subtilis WB800(pOALZ8) or B. subtilis WB800(pOAS-LZ8) cells were grown in LB medium or medium A, and the expression levels after 12 and

1042

YEH ET AL.


FIG. 2. Western blotting analysis of protein expression by B. subtilis DB104(pHSLZ) (A) or B. subtilis DB104(pHLZ) (B) cells cultured in LB medium. Lane M, broad-range molecular mass markers; lane 1, total cell lysate at 0 h; lane 2, medium fraction at 0 h; lanes 3 to 6, total cell lysate at 12, 24, 36, or 60 h, respectively; lanes 7 to 10, supernatant of B. subtilis DB104(pHSLZ) at 12, 24, 36, or 60 h, respectively; lane 11 (A), purified protein from total cell lysate; lane 11 (B) and lane 12 (A), purified recombinant LZ-8 protein from total cell lysate of B. subtilis DB104(pHLZ).

24 h of culture were analyzed. The productivity in medium A at 24 h for the two transformants was increased by 2.0- or 3.8-fold, respectively, over the level in LB medium (Fig. 3A), and the cell mass at 24 h was 2.7- or 5.3-fold higher than that of LB-cultured transformants (Fig. 3B). The expressed recombinant LZ-8s were purified by Ni-NTA affinity chromatography and confirmed by Western blot analysis and N-terminal-sequence analysis. An extra methionine was found at the N terminus of recombinant LZ-8 from B. subtilis WB800(pOALZ8), and the extra amino acids AEGSM were found before the N terminus of recombinant LZ-8 from B. subtilis WB800(pOASLZ8), as expected. These were added to allow translation initiation and construction for SP processing, respectively. The results showed that the SP was cleaved from recombinant LZ-8 produced by B. subtilis WB800(pOAS-LZ8). Extracellular expression of recombinant LZ-8 using SPYaB and expression improvement in Lactococcus lactis. The levels of expression of the recombinant LZ-8 by L. lactis(pZLZ) and L. lactis(pZSLZ) were analyzed. Both transformants were transcribed from SECS-4, which is identical to the consensus sequences of constitutive promoters of L. lactis (14). No expected bands were seen on the SDS-PAGE gels or Western blot analysis. A very weak band of the predicted size was seen on SDS-PAGE analysis of the Ni-NTA-purified supernatant proteins of L. lactis(pZSLZ), but not that of L. lactis(pZLZ). To improve and confirm the expression of recombinant LZ-8 in L. lactis and examine the expression of recombinant LZ-8 with or without the aid of the SP, its expression by pNZLZ and pNZSLZ in cells of L. lactis NZ9000 (21) and a cell wall protease-deficient host, L. lactis NZ9000 ⌬htrA (34), was examined. pNZLZ and pNZSLZ were transcribed from an effi-

cient PslpA1 promoter (3) and were expected to express recombinant LZ-8 in the cell cytosol and extracellular fraction, respectively. No expression of recombinant LZ-8 in the cell lysate or medium was observed for L. lactis NZ9000(pNZLZ) (data not shown). However, extracellular expression of recombinant LZ-8 was seen using the L. lactis NZ9000(pNZSLZ) transformant and was higher (1.3-fold) at the early stage (6 h of culture) using L. lactis NZ9000 ⌬htrA(pNZSLZ) (data not shown). To improve the production of recombinant LZ-8, various media were examined. Changing the medium from GM17 medium to 2GM17B medium resulted in a 2.5-fold increase in recombinant LZ-8 expression (Fig. 4A) and a 1.4-fold increase in cell mass (Fig. 4B) at 8 h of culture. The expressed recombinant LZ-8s were purified and confirmed by Western blot analysis and N-terminal sequence analysis. The extra amino acids AEGSM were found before the N terminus of recombinant LZ-8 from L. lactis NZ9000(pNZSLZ), showing that the SP was cleaved as effectively as in B. subtilis WB800(pOASLZ8). Homodimeric and secondary structure analysis of recombinant LZ-8s from the Bacillus subtilis transformants and Lactococcus lactis transformant. Using the improved culture conditions for B. subtilis WB800(pOA-LZ8), B. subtilis WB800 (pOAS-LZ8), and L. lactis NZ9000(pNZSLZ), extracellular recombinant LZ-8s were produced and purified by ammonium sulfate precipitation and native Ni-NTA affinity chromatography. After 12 h of culture of B. subtilis WB800(pOA-LZ8) or B. subtilis WB800(pOAS-LZ8) or 8 h of culture of L. lactis NZ9000 ⌬htrA(pNZSLZ), 17.5 mg, 13.2 mg, or 1.24 mg of extracellular recombinant LZ-8 was obtained from 1 liter of medium, re-

VOL. 74, 2008


1043

duction by the RAW 264.7 macrophage cell line, the two recombinant LZ-8s produced by B. subtilis sources stimulated high levels of production in a dose-dependent manner, while the recombinant LZ-8 from L. lactis NZ9000(pNZSLZ) was much less effective (Fig. 7). The immunomodulatory capacities of the three purified recombinant LZ-8s were determined by assaying IL-2 (Th1 cytokine) and IL-4 (Th2 cytokine) production by hPBMCs. The IL-2 (Th1 cytokine)/IL-4 (Th2 cytokine) ratios in the hPBMC cultures are shown in Fig. 8. The results showed that 25 ␮g/ml was the optimal concentration of all three recombinant LZ-8s for modulating the Th1/Th2 balance in hPBMC cultures. The recombinant LZ-8 from L. lactis NZ9000(pNZSLZ) achieved the highest IL-2/IL-4 ratio and significantly increased the IL2/IL-4 ratio at 12.5 and 50.0 ␮g/ml concentrations, while the recombinant LZ-8 from B. subtilis WB800(pOA-LZ8) showed Th1/Th2 modulator capacity over a range of 3.13 to 50.0 ␮g/ml concentrations and the recombinant LZ-8 from B. subtilis WB800(pOAS-LZ8) showed Th1/Th2 modulator capacity over a range of 6.25 to 50.0 ␮g/ml concentrations, broader ranges than that of recombinant LZ-8 from L. lactis NZ9000(pNZSLZ). DISCUSSION

FIG. 3. Levels of expression of recombinant LZ-8 (A) and cell masses (B) of B. subtilis WB800(pOA-LZ8) and B. subtilis WB800(pOASLZ8) cultured in various media. (A) Lane M, broad-range protein molecular mass marker; lanes 1 to 4 and lanes 5 to 8, supernatants of B. subtilis WB800(pOA-LZ8) and B. subtilis WB800(pOAS-LZ8) after 12 or 24 h of culture in LB medium or A medium, respectively. The upper and lower arrowheads indicate the recombinant LZ-8 proteins from B. subtilis WB800(pOAS-LZ8) and B. subtilis WB800(pOA-LZ8), respectively. (B) The closed and open circles represent, respectively, B. subtilis WB800(pOA-LZ8) and B. subtilis WB800(pOAS-LZ8) in LB medium, and the reverse triangles and closed triangles represent B. subtilis WB800(pOA-LZ8) and B. subtilis WB800(pOAS-LZ8) in A medium. OD600, optical density at 600 nm.

spectively. When the purified recombinant LZ-8s were subjected to homodimeric structure analysis using a cross-linker, all three recombinant LZ-8s showed a homodimeric structure (Fig. 5A). The CD analysis of the recombinant LZ-8 from L. lactis showed differences in the degree spectrum from recombinant LZ-8s from B. subtilis, indicating that the conformation of recombinant LZ-8s from B. subtilis and L. lactis might be slightly different (Fig. 5B). The three recombinant LZ-8s exhibit different capacities for immunomodulation of hPBMCs and abilities to stimulate TNF-␣ production by a macrophage cell line. The purified recombinant LZ-8s from three sources were examined for their effects on hPBMC proliferation. The recombinant LZ-8 from B. subtilis WB800(pOA-LZ8) at concentrations of 3.13 and 6.25 ␮g/ml caused a slight but significant increase in hPBMC numbers, while the recombinant LZ-8 from L. lactis NZ9000(pNZSLZ) at concentrations of 25.5 to 50.0 ␮g/ml caused an increase in hPBMC numbers that was not significant. The recombinant LZ-8s from B. subtilis sources induced bell-shaped proliferative response curves, but the recombinant LZ-8 from L. lactis did not (Fig. 6). When tested for their effects on TNF-␣ pro-

In attempts to express a eukaryote gene in prokaryote cells, the existence of introns in the gene and modifications, such as glycosylation, of the target protein must be considered. Examination of LZ-8 cDNA showed a small intron (61 nucleotides) in the 5⬘ untranslated region (35). LZ-8 contains 110 amino acids, has a calculated molecular mass of 12,420 Da, and has an acetyl group at the amino-terminal end (46). Native LZ-8 is a noncovalently associated homodimeric molecule (19). Recent studies of the FIPs FIP-fve (39) and FIP-gts (30) have revealed that the homodimeric structure is necessary for their immunomodulatory capacities. The N-terminal helix of FIP-gts is responsible for the homodimeric structure (30). Based on these observations, we attempted to express recombinant LZ-8 in B. subtilis and L. lactis using a novel recombinant LZ-8 gene designed based on the optimal codons for both B. subtilis and L. lactis. To facilitate purification, a six-histidine tag was incorporated at the C terminus of this novel recombinant LZ-8. To ensure extracellular expression, the cleavage sites of the fusion proteins recognized by signal peptidase and extra restriction sites needed for DNA cloning were considered and an AEGS sequence was therefore added in front of the N terminus of the extracellularly expressed recombinant LZ-8s. To ensure translation initiation, an extra M was added to the N terminus of cellularly expressed recombinant LZ-8s. The influence of the six-histidine tag and extra N-terminal sequence on the structure of recombinant LZ-8 was analyzed using Protein Homology/analogY Recognition Engine (http://www.sbg .bio.ic.ac.uk/phyre), and these additions were predicted not to affect the N-terminal helical structure, which is responsible for the homodimer structure. The novel recombinant LZ-8 gene, recombinant LZ-8, was synthesized by OE-PCR and expressed in the GRAS host B. subtilis and food-grade host L. lactis. For high-level extracellular expression of heterologous proteins, the factors to be considered are host engineering, codon optimization, promoter and Shine-Dalgarno sequence optimization, secretion optimization or coexpression of protein-fold-

1044

YEH ET AL.


FIG. 4. Expression levels of recombinant LZ-8 (A) and cell masses (B) of L. lactis NZ9000(pNZSLZ) cells cultured in various media. (A) Lane M, broad-range protein molecular mass markers; lanes 1 to 12, supernatants of L. lactis NZ9000(pNZSLZ) after 4, 6, or 8 h of culture in GM17, GM17B, 2GM17B, or 2NGM17B medium. (B) The circles, reverse triangles, squares, and diamonds represent growth in GM17, GM17B, 2GM17B, and 2NGM17B medium, respectively. OD600, optical density at 600 nm.

ing factors, and medium optimization. In this study, using the extremely efficient SECS-4 (51) to express recombinant LZ-8 in B. subtilis DB104, a three-protease-deficient host (18), did not achieve a high level of productivity; however, using the less-efficient SECS-2 (51) to express recombinant LZ-8 in the eight-protease-deficient B. subtilis WB800 (4) increased the expression. In our experience, the best SECS for a reporter gene may not be the best SECS for a target recombinant protein, as high-level expression of some heterologous proteins leads to modification of the SECS or mature protein gene (our unpublished data). The lethal effect of high-level expression might induce the host’s protection systems to modify the gene or expression control sequences. Surprisingly, recombinant LZ-8 was detected in the media of B. subtilis DB104(pHLZ) and B. subtilis WB800(pOA-LZ8). For extracellular expression, B. subtilis, which, like other grampositive bacteria, lacks an outer membrane, is superior to gram-negative bacteria, such as E. coli. Four protein secretion pathways have been identified in B. subtilis in a genome-wide survey (47). Of these, the sec-dependent pathway is the most widely used (47). The efficiency of SP processing depends on the fusion cleavage site (54). SPYaB is a sec-dependent SP, and its length (26 amino acids) and fusion cleavage site (SSIAQA 2AEGSM) are predicted to be preferred by B. subtilis (47). Using B. subtilis DB104(pHSLZ) and B. subtilis WB800(pOASLZ8), recombinant LZ-8 was indeed expressed extracellularly and processed by signal peptidase at the predicted site. Interestingly, with B. subtilis DB104(pHLZ) or B. subtilis WB800(pOALZ8), recombinant LZ-8 was expressed extracellularly without the aid of an SP. The N-terminal amino acid sequence of the

recombinant LZ-8s produced by these two transformants was MDTAL. The extra methionine was incorporated to ensure translation initiation and was not removed. It is possible that the extracellular expression of recombinant LZ-8 by B. subtilis DB104(pHLZ) and B. subtilis WB800(pOA-LZ8) was due to leakage caused by loss of integrity of the cell membrane. Because the recombinant LZ-8s from B. subtilis DB104(pHSLZ) and B. subtilis WB800(pOAS-LZ8) were secreted in a secdependent manner, secretion was increased by the increasing efficiency of major signal peptidases (47). Meanwhile, the extracellular recombinant LZ-8s produced by B. subtilis DB104(pHLZ) and B. subtilis WB800(pOA-LZ8), which lacked possible secretion signals, were suddenly increased at 60 h and 12 h of culture, respectively, suggesting a leakage due to loss of cell integrity under high-level-expression stress. The secretion machinery protects heterologous proteins from degradation by cellular proteases (5). In these overexpression conditions, the recombinant LZ-8 of B. subtilis WB800(pOA-LZ8) was not degraded by cellular proteases and possibly was folded in the cell cytosolic environment. The expression of recombinant LZ-8 by B. subtilis WB800(pOA-LZ8) and B. subtilis WB800(pOAS-LZ8) was then improved by culturing in various media. The optimization of medium components can result in high-level recombinant protein production (40). Medium A improved the recombinant LZ-8 productivity of B. subtilis WB800(pOA-LZ8) and B. subtilis WB800(pOAS-LZ8) (Fig. 4A) and increased the cell mass in the 24-h cultures (Fig. 4B). A decrease, followed by an increase, in cell mass was observed at 8 to 15 h with both transformants. It is possible that the death of some cells leads

VOL. 74, 2008


1045

FIG. 6. Effects of recombinant LZ-8 (rLZ-8) from B. subtilis WB800 (pOAS-LZ8), B. subtilis WB800(pOA-LZ8), and L. lactis NZ9000 (pNZSLZ) on the viability or proliferation of hPBMCs. hPBMCs were adjusted to 1.6 ⫻ 106 cells/ml in TCM medium and treated with various recombinant LZ-8s at 0 to 50 ␮g/ml final concentrations or PBS buffer as control and incubated at 37°C for 72 h. The dashed line shows the results for control hPBMCs without recombinant LZ-8 treatment. The circles, inverted triangles, and squares represent the recombinant LZ-8, proteins from B. subtilis WB800(pOA-LZ8), B. subtilis WB(pOAS-LZ8), and L. lactis NZ9000(pNZSLZ), respectively. The asterisks indicate significant differences against the control at a P value of ⬍0.05. OD600, optical density at 600 nm.

FIG. 5. Tricine-SDS-PAGE analysis of the homodimeric structure (A) and CD spectrum (B) of purified recombinant LZ-8 from B. subtilis WB800(pOAS-LZ8), B. subtilis WB800(pOA-LZ8), or L. lactis NZ9000(pNZSLZ). (A) Lane M, broad-range protein molecular mass markers; lanes 1 and 2, recombinant LZ-8 from B. subtilis WB800(pOA-LZ8) without (lane 1) or with (lane 2) cross-linker; lanes 3 and 4, recombinant LZ-8 from B. subtilis WB800(pOAS-LZ8) without (lane 3) or with (lane 4) cross-linker; lanes 5 and 6, recombinant LZ-8 from L. lactis NZ9000(pNZSLZ) without (lane 5) or with (lane 6) cross-linker. (B) Solid line represents recombinant LZ-8 from B. subtilis WB800(pOA-LZ8), dotted line represents recombinant LZ-8 from B. subtilis WB800(pOAS-LZ8), and dashed line represents recombinant LZ-8 from L. lactis NZ9000(pNZSLZ). OD600, optical density at 600 nm.

to the release of factors, such as phosphate regulator peptides (24), that may affect cell density and cell growth. The high cell density led to high productivity. By using L. lactis NZ9000(pNZSLZ), in which recombinant LZ-8 was transcribed from an efficient PslpA1 promoter (31), extracellular expression of recombinant LZ-8 was achieved. By using a cell wall protease-deficient host, Lactococcus lactis NZ9000 ⌬htrA, the extracellular production of recombinant LZ-8 was increased 1.3-fold at the early stage (6 h of culture). Although the artificial SECS-4 was identical to the consensus sequence of L. lactis strong promoters (14), it was not efficient in L. lactis. There might be other, dominant factors that influence transcription efficiency in L. lactis. In the PslpA1 promoter, the 5⬘ untranslated leader sequence of the slpA gene contributes to mRNA stabilization by producing a 5⬘ stem and loop structure (37). In this work, the PCR primers amplified the PslpA1 promoter region, including the 5⬘ untranslated leader sequence, and high-level extracellular production of recombi-

nant LZ-8 was achieved. This suggests that mRNA stability is more important than promoter strength in recombinant LZ-8 expression in L. lactis. Extra amino acids (AEGSD) were found in front of the N terminus of the purified recombinant

FIG. 7. Effects of recombinant LZ-8 (rLZ-8) from B. subtilis WB800 (pOAS-LZ8), B. subtilis WB800(pOA-LZ8), or L. lactis NZ9000 (pNZSLZ) on TNF-␣ secretion by the RAW 264.7 macrophage line. RAW 264.7 cells were cultured at a density of 1 ⫻ 105 cells without or with recombinant LZ-8 (0 to 50 ␮g/ml final concentration) at 37°C for 72 h, and then TNF-␣ was assayed using ELISA kits. The circles, reverse triangles, and squares represent the recombinant LZ-8 from B. subtilis WB800(pOA-LZ8), B. subtilis WB800(pOAS-LZ8), and L. lactis NZ9000(pNZSLZ), respectively. Differences between sample results were determined by Duncan’s multiple range test at a P value of ⬍0.05. The same letter indicates no significant difference between samples.

1046

YEH ET AL.

FIG. 8. Effects of purified recombinant LZ-8s (rLZ-8) from B. subtilis WB800(pOAS-LZ8), B. subtilis WB800(pOA-LZ8), and L. lactis(pNZSLZ) on the IL-2/IL-4 ratio of hPBMC. The blank bar represents control hPBMCs without recombinant LZ-8 treatment; the gray bars, black bars, and hatched bars represent purified recombinant LZ-8s from B. subtilis WB800(pOA-LZ8), B. subtilis WB800(pOAS-LZ8), and L. lactis(pNZSLZ), respectively. Differences among sample results were determined by Duncan’s multiple range tests at a P value of ⬍0.05. The same letter indicates no significant difference among samples.

LZ-8 from L. lactis NZ9000(pNZSLZ), indicating that the SP was cleaved as effectively as in B. subtilis. Although we have not yet examined secretion efficiency, the SPYaB from the heterologous Bacillus YaB source functioned well in L. lactis. In contrast to the extracellular leaky expression of recombinant LZ-8 seen in B. subtilis, no recombinant LZ-8 was detected in either the cell lysate or medium of the L. lactis NZ9000(pNZLZ) and L. lactis NZ9000 ⌬htrA(pNZLZ) transformants. Since L. lactis lacks extracellular proteases (24), recombinant LZ-8 was possibly recognized as a misfolded protein and degraded in the cell lysate by cellular degrading systems, such as the major housekeeping protease CIP complex, or other cellular components in L. lactis (26). With culture in 2GM17B medium, L. lactis NZ9000(pNZSLZ) exhibited the highest recombinant LZ-8 productivity, the cell mass of L. lactis NZ9000(pNZSLZ) being 1.4-fold higher in this medium than in GM17 or 2NGM17B medium and 2.0-fold higher than in GM17B medium (Fig. 4B). Culturing under pH-controlled conditions lessens acidification, improves energy utilization, and leads to higher biomass production of acid-tolerant fermentative bacteria (8, 45). In this study, the use of a pH-controlled GM17 medium (GM17B) resulted in 1.6-fold-higher recombinant LZ-8 productivity but a lower biomass at 6 h than the use of the nonbuffered GM17 medium. In addition, enriched buffer media (2GM17B and 2NGM17B) resulted in higher recombinant LZ-8 productivities at 8 h than GM17B medium. With double the concentration of glucose in the medium, the use of the pH-controlled medium 2GM17B significantly increased the biomass and recombinant LZ-8 productivity, while doubling the concentrations of peptone and yeast extract in the pH-controlled medium 2NGM17B did not influence the biomass and only slightly increased recombinant LZ-8 productivity at 6 h compared to the biomass and recombinant LZ-8 productivity in GM17B medium. The results sug-


gest that glucose, rather than the N source, contributes to the biomass increase in pH-controlled conditions in L. lactis and that the glucose concentration must be high enough to support the growth of L. lactis in a pH-controlled medium. Native LZ-8 purified from Ganoderma lucidium has potent effects on the immune system, such as a mitogenic effect on human peripheral blood lymphocytes and a T-cell-activator effect mediated via cytokine regulation (9). Recently, many immunomodulatory proteins from fungal sources, known as FIPs, have been identified. FIP-gts, FIP-fve, and FIP-vvo were purified from Ganoderma tsugae (30), the edible mushroom Flammulina velutipes (22), and Volvariella volvacea (13), respectively. FIP-gts, with an amino acid sequence identical to that of LZ-8, was functional when expressed in E. coli (30) and inhibited the transcriptional regulation of telomerase in the A549 human lung adenocarcinoma cell line (28). FIP-fve was cloned and functional when expressed in E. coli (23). Oral administration of FIP-fve has a Th1-skewing effect on the development of the allergen-specific immune response (12). The production of FIPs from natural resources is costly and time consuming. The reported recombinant FIPs were mostly produced in E. coli, which might lead to contamination with endotoxins and immunogenic lipopolysaccharides. The use of GRAS or food-grade hosts to produce recombinant FIPs is highly recommended for FIPs for oral administration. In this study, the recombinant FIP, recombinant LZ-8, was produced extracellularly in the GRAS host B. subtilis and food-grade host L. lactis. The recombinant LZ-8s from three sources were homodimeric in structure (Fig. 5A), in agreement with the original design and prediction. The effect of purified recombinant LZ-8s on hPBMC proliferation was examined by MTT assay. The recombinant LZ-8s from B. subtilis sources induced bellshaped proliferation response curves at concentrations from 0 to 25 ␮g/ml, comparable to the effect of wild LZ-8 on the human peripheral blood lymphocytes at concentrations from 0.1 to 10 ␮g/ml (9, 50). The recombinant LZ-8 from L. lactis did not exhibit a bell-shaped proliferation response curve (Fig. 6). The recombinant LZ-8 from B. subtilis WB800(pOA-LZ8) significantly increased hPBMC proliferation at concentrations of 3.13 and 6.25 ␮g/ml. Although the recombinant LZ-8s from B. subtilis WB800(pOAS-LZ8) and L. lactis NZ9000(pNZSLZ) increased hPBMC proliferation at concentrations of 3.13 to 12.5 ␮g/ml and 12.5 to 50.0 ␮g/ml, respectively, the effect was not significant. The immunomodulatory capabilities of the three recombinant LZ-8s were examined. It is well recognized that immune defense against infection by pathogens is mediated by helper CD4⫹ T (Th) cells (41). Two functionally distinct subsets of cytokines are activated, and the Th cells differentiate into Th1 cells and Th2 cells. Th1 cells secrete the cytokines IL-2, gamma interferon, and TNF-␤, which are linked to cell-mediated immunity and eliminate intracellular pathogens via macrophage activation, while Th2 cells secrete the cytokines IL-4, IL-5, IL-10, and IL-13, which affect humoral immunity and responses to persistent antigens and allergens (41, 44). The two helper subsets also cross-regulate each other, so the balance between Th1 and Th2 cytokines can determine whether the immune response is appropriate or not (41). Overproduction of Th1 cytokines has been implicated in delayed-type hyper-

VOL. 74, 2008


sensitivity reactions and autoimmune diseases. Th2 cytokines activate mast cells and lead to allergic and inflammatory conditions. Thus, immunomodulation of the balance between Th1 and Th2 cells might reduce the incidence of allergic diseases. In this study, the immunomodulatory capacities of recombinant LZ-8s from three recombinant resources were analyzed by the IL-2/IL-4 ratio. All three significantly stimulated cytokine production (data not shown). Compared to the results for the PBS control, all three recombinant LZ-8s had different immunomodulatory capacities to direct Th1 cytokine production (Fig. 8). For all three recombinant LZ-8s, the concentration of 25 ␮g/ml resulted in the highest IL-2/IL-4 ratio in hPBMCs. At the higher concentrations (25 ␮g/ml and 50 ␮g/ ml), the IL-2/IL-4 ratio did not differ significantly for the three different recombinant LZ-8s. At lower concentrations (3.125 to 6.25 ␮g/ml), recombinant LZ-8 from L. lactis NZ9000(pNZSLZ) did not significantly affect the IL-2/IL-4 ratio, whereas 3.125 to 50.0 ␮g/ml of recombinant LZ-8 from B. subtilis WB800(pOALZ8) or 6.25 to 50 ␮g/ml of recombinant LZ-8 from B. subtilis WB800(pOA-LZ8) resulted in a significant increase in the IL-2/IL-4 ratio. This suggested that recombinant LZ-8s produced by B. subtilis and L. lactis might be used as immunomodulatory agents in the immunoprophylaxis of food allergy and other allergic diseases. Interestingly, the three recombinant LZ-8s exhibited different immunomodulatory capacities. The N-terminal amphipathic ␣-helix is responsible for the homodimeric structure and plays an important role in the immunomodulatory activities of FIPs (30, 39). Mutants of FIP-gts lost the amphipathic characteristics of the N-terminal domain, the ability to form dimers, and immunomodulatory activity (30). Thus, the stability of the N-terminal amphipathic domain may influence the immunomodulatory activities of FIPs. In this study, the N termini of the recombinant LZ-8s were different from that of wild-type LZ-8 because of cloning requirements. The initiation methionine of the wild-type LZ-8 was removed and modified by an acetyl group (acetyl SETA) at the N terminus (46). The N termini of the recombinant LZ-8s from B. subtilis WB800(pOAS-LZ8) and L. lactis NZ9000(pNZSLZ) had an extra AEGSM (AEGSMSETA) with carboxyl, hydroxyl, and thioether side chains added, while the recombinant LZ-8 from B. subtilis WB800(pOA-LZ8) had only an extra methionine (MSETA) added. These extra amino acids did not affect the formation of the homodimeric structure (Fig. 5A). The CD spectra of the recombinant LZ-8s from B. subtilis were similar, while the recombinant LZ-8 produced by L. lactis showed a similar but variable degree spectrum, indicating a slight conformation difference. Despite having an identical primary sequence, the extracellularly expressed recombinant LZ-8s from B. subtilis WB800(pOAS-LZ8) and L. lactis NZ9000(pNZSLZ) had different immunomodulatory capacities. Two possible reasons, the protein modifications or conformation variability, were considered. The results of mass spectrophotometry confirmed that the two recombinant LZ-8s were the same, thus excluding the possibility of protein modification. In the CD analysis, the recombinant LZ-8 produced by L. lactis showed a similar but more-variable degree spectrum, indicating a difference between conformations. A theory of “protein memory” has been proposed, in which an identical polypeptide can fold into an

1047

altered conformation with a different secondary structure, stability, and specificities with the guidance of mutated intramolecular chaperones (IMCs) (44). The structures of the N-domain, H-domain, C-domain, and SP cleavage site of SPYaB indicate that it is sec-dependent (7), indicating that the recombinant LZ-8s were secreted via the sec-dependent pathway, which directs the majority of secretory proteins into the growth medium in B. subtilis and L. lactis (7, 8). The well-characterized Sec-SRP pathway can be divided into three stages: targeting, translocation, and folding and release (7). Within the cell, the protein is in an unfolded state and bound to IMCs, and it is then translocated by the membrane-integrated translocation component before being folded in the extracellular environment. The IMCs of L. lactis have not been extensively studied. A recent article (26) reported that L. lactis is capable of secreting proteins with molecular masses from 6 kDa to 105 kDa and that the conformation, rather than the size, affects protein secretion efficiency in L. lactis, but no such information is available for the well-studied B. subtilis. The different features of the IMCs and translocation components of B. subtilis and L. lactis might affect the structure-function relationships of the recombinant LZ-8s with their identical primary sequences that are produced by B. subtilis and L. lactis. The recombinant LZ-8 produced by B. subtilis WB800(pOALZ8) cells induced a higher IL-2/IL-4 ratio over a broader concentration range (3.13 to 50.0 ␮g/ml) than the recombinant LZ-8 produced by B. subtilis WB800(pOAS-LZ8) cells, which induced a lower IL-2/IL-4 ratio over a concentration range from 6.25 to 50.0 ␮g/ml. In the CD analysis, the recombinant LZ-8s produced by B. subtilis showed degree spectra that were similar to but slightly less variable than that of L. lactis. The differences between immunomodulatory capacities might due to the extra N-terminal amino acid sequences affecting the binding of the receptor or to slight differences in conformation that occurred during the protein-folding process. The recombinant LZ-8 of B. subtilis WB800(pOA-LZ8) was possibly folded in the cytosol environment, while the recombinant LZ-8 of B. subtilis WB800(pOASLZ8) was folded with the guidance of IMCs (44). However, compared to the immunomodulatory capacity and CD spectrum of the recombinant LZ-8 from L. lactis, the immunomodulatory capacities and CD spectra of the recombinant LZ-8s from B. subtilis were similar. For antitumor activity, recombinant LZ-8 stimulated the production of TNF-␣, which induces apoptosis and differentiation in treated leukemic cells (52). The polysaccharide fraction (PS-G) isolated from the Ganoderma lucidium fresh fruit body was reported to have antitumor effects, not from PS-G itself, but rather, the conditioned medium from PS-G-activated mononuclear cells was highly antiproliferative (52). LZ-8 might contribute to the antitumor activity of Ganoderma lucidium, as is the case for FIP-gts in Ganoderma tsugae (28). The extracellularly expressed recombinant LZ-8 produced by B. subtilis WB800(pOAS-LZ8) resulted in the highest stimulation of TNF-␣ production. The structure-function relationship and mechanism of the recombinant LZ-8 in the antitumor activity still need to be investigated. In conclusion, the codon-optimized Ganoderma lucidium immunomodulatory protein LZ-8 gene, recombinant LZ-8, was expressed extracellularly in the GRAS host Bacillus subtilis with or without the aid of SPYaB. The food-grade host Lacto-

1048

YEH ET AL.


coccus lactis expressed recombinant LZ-8 extracellularly with the aid of SPYaB. Three recombinant LZ-8s exhibited potent antitumor activities and different capacities for modulating Th1 and Th2 cytokines. The recombinant LZ-8 may be useful in immunoprophylaxis for food allergies and other allergic diseases or cancer therapy. The recombinant LZ-8 purified from B. subtilis can be used as a food supplement. Optimization of expression and secretion efficiency in food-grade L. lactis will provide a useful source of recombinant LZ-8 for safe use as an orally administered agent. ACKNOWLEDGMENTS This work was supported in part by a grant (93AS-4.1.4-AD-U1) from the Agricultural Council, Taiwan, Republic of China. We thank J. Y. Lin for his kindness in providing technical support for the MTT and cytokine assays. H. C. Hung and K. L. Su are appreciated for their kindness in providing the CD spectropolarimeter and technical support. REFERENCES 1. Bermu ´dez-Humara ´n, L. G., P. Langella, A. Gruss, R. Montes de Oca-Luna, and Y. Le Loir. 2002. Production of human papillomavirus type 16 E7 protein in Lactococcus lactis. Appl. Environ. Microbiol. 68:917–922. 2. Blatny, J. M., H. Ertesvag, I. F. Nes, and S. Valla. 2003. Heterologous gene expression in Lactococcus lactis: expression of the Azotobacter vinelandii algE6 gene product displaying mannuronan C-5 epimerase activity. FEMS Microbiol. Lett. 227:229–235. 3. Boot, H. J., C. P. Kolen, F. J. Andreadaki, R. J. Leer, and P. H. Pouwels. 1996. The Lactobacillus acidophilus S-layer protein gene expression site comprises two consensus promoter sequences, one of which directs transcription of stable mRNA. J. Bacteriol. 178:5388–5394. 4. Braaz, R., S. L. Wong, and D. Jendrossek. 2002. Production of PHA depolymerase A (PhaZ5) from Paucimonas lemoignei in Bacillus subtilis. FEMS Microbiol. Lett. 209:237–241. 5. Choi, J. H., and S. Y. Lee. 2004. Secretory and extracellular production of recombinant proteins using Escherichia coli. Appl. Microbiol. Biotechnol. 64:625–635. 6. de Vos, W. M. 1999. Gene expression systems for lactic acid bacteria. Curr. Opin. Microbiol. 2:289–295. 7. Fu, L. L., Z. R. Xu, W. F. Li, J. B. Shuai, P. Lu, and C. X. Hu. 2007. Protein secretion pathways in Bacillus subtilis: implication for optimization of heterologous protein secretion. Biotechnol. Adv. 25:1–12. 8. Fu, R. Y., J. Chen, and Y. Li. 2005. Heterologous leaky production of transglutaminase in Lactococcus lactis significantly enhances the growth performance of the host. Appl. Environ. Microbiol. 71:8911–8919. 9. Haak-Frendscho, M., K. Kino, T. Sone, and P. Jardieu. 1993. Ling Zhi-8: a novel T cell mitogen induces cytokine production and upregulation of ICAM-1 expression. Cell. Immunol. 150:101–113. 10. Harwood, C. R. 1992. Bacillus subtilis and its relatives: molecular biological and industrial workhorses. Trends Biotechnol. 10:247–256. 11. Hols, P., M. Kleerebezem, A. N. Schanck, T. Ferain, J. Hugenholtz, J. Delcour, and W. M. de Vos. 1999. Conversion of Lactococcus lactis from homolactic to homoalanine fermentation through metabolic engineering. Nat. Biotechnol. 17:588–592. 12. Hsieh, K. Y., C. I. Hsu, J. Y. Lin, C. C. Tsai, and R. H. Lin. 2003. Oral administration of an edible-mushroom-derived protein inhibits the development of food-allergic reactions in mice. Clin. Exp. Allergy 33:1595–1602. 13. Hsu, H. C., C. I. Hsu, R. H. Lin, C. L. Kao, and J. Y. Lin. 1997. Fip-vvo, a new fungal immunomodulatory protein isolated from Volvariella volvacea. Biochem. J. 323:557–565. 14. Jensen, P. R., and K. Hammer. 1998. Artificial promoters for metabolic optimization. Biotechnol. Bioeng. 58:191–195. 15. Reference deleted. 16. Kaltwasser, M., T. Wiegert, and W. Schumann. 2002. Construction and application of epitope- and green fluorescent protein-tagging integration vectors for Bacillus subtilis. Appl. Environ. Microbiol. 68:2624–2628. 17. Kang, I. S., J. J. Wang, J. C. H. Shih, and T. C. Lanier. 2004. Extracellular production of a functional soy cystatin by Bacillus subtilis. J. Agric. Food Chem. 52:5052–5056. 18. Kawamura, F., and R. H. Doi. 1984. Construction of a Bacillus subtilis double mutant deficient in extracellular alkaline and neutral protease. J. Bacteriol. 160:442–444. 19. Kino, K., A. Yamashita, K. Yamaoka, J. Watanabe, S. Tanaka, K. Ko, K. Shimizu, and H. Tsunoo. 1989. Isolation and characterization of a new immunomodulatory protein, Ling Zhi-8 (LZ-8), from Ganoderma lucidium. J. Biol. Chem. 264:472–478.

20. Kino, K., T. Mizumoto, T. Sone, T. Yamaji, J. Watanabe, A. Yamashita, K. Yamaoka, K. Shimizu, K. Ko, and H. Tsunoo. 1990. An immunomodulatory protein, Ling Zhi-8 (LZ8), prevents insulitis in non-obese diabetic mice. Diabetologia 33:713–718. 21. Kleerebezem, M., M. M. Beerhuyzen, E. E. Vaughan, W. M. de Vos, and O. P. Kuipers. 1997. Controlled gene expression systems for lactic acid bacteria: transferable nisin-inducible expression cassettes for Lactococcus, Leuconostoc, and Lactobacillus spp. Appl. Environ. Microbiol. 63:4581–4584. 22. Ko, J. L., C. I. Hsu, R. H. Lin, C. L. Kao, and J. Y. Lin. 1995. A new fungal immunomodulatory protein, FIP-fve, isolated from the edible mushroom Flammulina velutipes, and its complete amino acid sequence. Eur. J. Biochem. 228:244–249. 23. Ko, J. L., S. J. Lin, C. I. Hsu, C. L. Kao, and J. Y. Lin. 1997. Molecular cloning and expression of a fungal immunomodulatory protein, FIP-fve, from Flammulina velutipes. J. Formos. Med. Assoc. 96:517–524. 24. Koetje, E. J., A. Hajdo-Milasinovic, R. Kieweit, S. A. Born, and H. Tjalsma. 2003. A plasmid-borne Rap-Phr system of Bacillus subtilis can mediate celldensity controlled production of extracellular proteases. Microbiology 149: 19–28. 25. Kunji, E. R., D. J. Slotboom, and B. Poolman. 2003. Lactococcus lactis as host for overproduction of functional membrane proteins. Biochim. Biophys. Acta 1610:97–108. 26. Le Loir, Y., V. Azevedo, S. C. Oliveira, D. A. Freitas, A. Miyoshi, L. G. Bermu ´dez-Humara ´n, S. Nouaille, L. A. Ribeiro, S. Leclercq, J. E. Gabriel, V. D. Guimaraes, M. N. Oliveira, C. Charlier, M. Gautier, and P. Langella. 2005. Protein secretion in Lactococcus lactis: an efficient way to increase the overall heterologous protein production. Microb. Cell Fact. 4:2–10. 27. Leroy, F., and L. Devuyst. 2004. Lactic acid bacteria as functional starter cultures for the food fermentation industry. Trend. Food Sci. Technol. 15: 67–78. 28. Liao, C. H., Y. M. Hsiao, C. P. Hsu, M. Y. Lin, J. C. Wang, Y. L. Huang, and J. L. Ko. 2006. Transcriptionally mediated inhibition of telomerase of fungal immunomodulatory protein from Ganoderma tsugae in A549 human lung adenocarcinoma cell line. Mol. Carcinog. 45:220–229. 29. Lin, J. Y., Y. S. Lai, C. J. Liu, and A. R. Wu. 2007. Effects of lotus plumule supplementation before and following systemic administration of lipopolysaccharide on the splenocyte responses of BALB/c mice. Food Chem. Toxicol. 45:486–493. 30. Lin, W. H., C. H. Hung, C. I. Hsu, and J. Y. Lin. 1997. Dimerization of the N-terminal amphipathic alpha-helix domain of the fungal immunomodulatory protein from Ganoderma tsugae (Fip-gts) defined by a yeast two-hybrid system and site-directed mutagenesis. J. Biol. Chem. 272:20044–20048. 31. Liu, C. F., H. J. Peng, C. K. Yeh, C. H. Su, and C. M. Yeh. 2006. Study on the food grade vectors for Lactobacillus paracasei. Taiwan. J. Agric. Chem. Food Sci. 44:351–363. 32. Mierau, I., P. Leij, I. van Swam, B. Blommestein, E. Floris, J. Mond, and E. J. Smid. 2005. Industrial-scale production and purification of a heterologous protein in Lactococcus lactis using the nisin-controlled gene expression system NICE: the case of lysostaphin. Microb. Cell Fact. 4:15–23. 33. Miyasaka, N., H. Inoue, T. Totsuka, R. Koike, K. Kino, and H. Tsunoo. 1992. An immunomodulatory protein, Ling Zhi-8, facilitates cellular interaction through modulation of adhesion molecules. Biochem. Biophys. Res. Commun. 186:385–390. 34. Miyoshi, A., I. Poquet, V. Azevedo, J. Commissaire, L. Bermudez-Humaran, E. Domakova, Y. Le Loir, S. C. Oliveira, A. Gruss, and P. Langella. 2002. Controlled production of stable heterologous proteins in Lactococcus lactis. Appl. Environ. Microbiol. 68:3141–3146. 35. Murasugi, A., S. Tanaka, N. Komiyama, N. Iwata, K. Kino, H. Tsunoo, and S. Sakuma. 1991. Molecular cloning of a cDNA and a gene encoding an immunomodulatory protein, Ling Zhi-8, from a fungus, Ganoderma lucidium. J. Biol. Chem. 266:2486–2493. 36. Nagarajan, V. 1990. System for secretion of heterologous proteins in Bacillus subtilis, p. 214–223. In D. V. Goeddel (ed.), Gene expression technology. Academic Press, London, United Kingdom. 37. Narita, J., S. Ishida, K. Okano, S. Kimura, H. Fukuda, and A. Kondo. 2006. Improvement of protein production in lactic acid bacteria using 5⬘-untranslated leader sequence of slpA from Lactobacillus acidophilus. Appl. Microbiol. Biotechnol. 73:366–373. 38. Nouaille, S., L. A. Ribeiro, A. Miyoshi, D. Pontes, Y. Le Loir, S. C. Oliveira, P. Langella, and V. Azevedo. 2003. Heterologous protein production and delivery systems for Lactococcus lactis. Genet. Mol. Res. 2:102–111. 39. Paaventhan, P., J. S. Joseph, S. V. Seow, S. Vaday, H. Robinson, K. Y. Chua, and P. R. Kolatkar. 2003. A 1.7Å structure of Fve, a member of the new fungal immunomodulatory protein family. J. Mol. Biol. 332:461–470. 40. Park, K., and K. F. Reardon. 1996. Medium optimization for recombinant protein production by Bacillus subtilis. Biotechnol. Lett. 18:737–740. 41. Rengarajan, J., S. Szabo, and L. Glimcher. 2000. Transcriptional regulation of Th1/Th2 polarization. Immunol. Today 21:479–483. 42. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory manual, p. A3.1–A3.10. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 43. Scha ¨gger, H., and G. I. von Jagow. 1987. Tricine-sodium dodecyl sulfate-

VOL. 74, 2008

44. 45.

46.

47.

48.

49.

50.

51.


polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal. Biochem. 166:368–379. Shinde, U., J. J. Liu, and M. Inouye. 1997. Protein memory through altered folding mediated by intramolecular chaperones. Nature 389:520–522. Sriraman, H., and G. Jayaraman. 2006. Enhancement of recombinant streptokinase production in Lactococcus lactis by suppression of acid tolerance response. Appl. Microbiol. Biotechnol. 72:1202–1209. Tanaka, S., K. Ko, K. Kino, K. Tsuchiya, A. Yamashita, A. Murasugi, S. Sakuma, and H. Tsunoo. 1989. Complete amino acid sequence of an immunomodulatory protein, Ling Zhi-8 (LZ8). J. Biol. Chem. 264:16372–16377. Tjalsma, H., A. Bolhuis, J. D. H. Jongbloed, S. Bron, and J. M. van Dijl. 2000. Signal peptide-dependent protein transport in Bacillus subtilis: a genome-based survey of the secretome. Microbiol. Mol. Biol. Rev. 64:515–547. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350–4354. van der Hem, L. G., J. A. van der Vliet, K. Kino, A. J. Hoitsma, and W. J. M. Tax. 1996. Ling Zhi-8: a fungal protein with immunomodulatory effects. Transplant. Proc. 28:958–959. van der Hem, L. G., J. A. van der Vliet, C. Frans, M. Bocken, K. Kino, A. J. Hoitsma, and W. J. M. Tax. 1995. Ling Zhi-8: studies of a new immunomodulating agent. Transplantation 60:438–443. Wang, J. P., C. M. Yeh, and Y. C. Tsai. 2006. Improved production of

52.

53.

54.

55. 56.

57. 58.

1049

recombinant subtilisin YaB by engineering the expression control sequences in Bacillus subtilis. J. Agric. Food Chem. 54:9405–9410. Wang, S. Y., M. L. Hsu, H. C. Hsu, S. S. Lee, M. S. Shiao, and C. K. Ho. 1997. The anti-tumor effect of Ganoderma lucidium is mediated by cytokines released from activated macrophages and T lymphocytes. Int. J. Cancer 70: 699–705. Westers, L., D. S. Dijkstra, H. Westers, J. M. van Dijl, and W. J. Quax. 2006. Secretion of functional human interleukin-3 from Bacillus subtilis. J. Biotechnol. 123:211–224. Wetmore, D. R., S. L. Wong, and R. S. Roche. 1994. The efficiency of processing and secretion of the thermolysin-like neutral protease from Bacillus cereus does not require the whole prosequence, but does depend on the nature of the amino acid sequence in the region of the cleavage site. Mol. Microbiol. 12:747–759. Wong, S. L. 1995. Advances in the use of Bacillus subtilis for the expression and secretion of heterologous proteins. Curr. Opin. Biotechnol. 6:517–522. Yeh, C. M., J. P. Wang, and F. S. Su. 2005. Improved protocol for Bacillus subtilis DB104 applicable in ligation mixtures direct electro-transformation. Taiwan. J. Agric. Chem. Food Sci. 43:325–333. Yeh, C. M., and Y. T. Chou. 2006. Study on the cell surface display system of lactic acid bacteria. Taiwan. J. Agric. Chem. Food Sci. 44:283–295. Yeh, C. M., J. P. Wang, and F. S. Su. 2007. Extracellular production of a novel ice structuring protein by Bacillus subtilis: a case of recombinant food peptide additive production. Food Biotechnol. 21:119–128.