Introduction

The pharmaceutical industry is applying recombinant protein technology to satisfy the need to find a system to produce drugs on a large scale but at a low production cost. The use of transgenic plants as a system for the expression of relevant antigens has been widely assessed, and the general conclusion drawn is that the large-scale production of such antigens is associated with a relatively low cost and less complex production system (Daniell et al. 2009; Ramessar et al. 2009). In addition, results from studies on animals under experimental conditions and from clinical assays in humans indicate that certain antigens derived from transgenic plants act as prophylactic agents and can be used to treat certain diseases (Tacket et al. 2000; Walmsley and Arntzen 2000). Hence, these plant-produced recombinant proteins have been proposed as an alternative production system to support public health programs aimed at controlling diseases that cause devastating problems in the population (Tacket et al. 2000), such as rheumatoid arthritis (RA). In Mexico, RA is one of the most common chronic diseases, with important physical, social, economic and psychological consequences. Statistics from the health sector in 2005 reported that 1.5 million people suffer arthritic problems (http://www.issste.gob.mx/website/comunicados/boletines/2005/mayo2005.html).

In recent decades, extensive research efforts have been directed towards determining the primary etiological agent causing RA. However, the causes are still a mystery, since it is thought that multiple factors, such as heredity, hormonal changes, environment, and/or infection with another viral or bacterial pathogen, may be involved (Lunardi et al. 2000; Moctezuma 2002; Lopes-Silva et al. 2009). Molecular studies have reported the existence of certain molecules that may themselves be involved in the development and modulation of RA, such as the type II collagen molecule and the glycoprotein and human cartilage protein CH65, which are expressed in the cartilage of arthritic patients. However, there is strong evidence indicating the minimal involvement of these proteins in the development of the disease (Rudolphi et al. 1997).

It has been reported that a number of autoimmune diseases occur after primary infection with a pathogen, with the pathogen producing a molecule that co-reacts immunologically with another molecule present in the tissues of the host, causing the destruction of these tissues (Life et al. 1993; Matzinger 1994; Chen et al. 1999; Kohm et al. 2003). This hypothesis has been strongly supported by results from studies on arthritis induced by Mycobacterium tuberculosis attenuated by heat which show the activation of T cells and antibodies that recognize an antigen of the 65-KDa mycobacterial protein, which subsequently increases its expression in the process of inflammation and heat stress (van Eden 1988; Blass et al. 1999; Sewell et al. 2002; Lopes-Silva et al. 2009).

Heat stress proteins (HSPs) show a high degree of evolutionary conservation in eukaryotes and prokaryotes (Brown and Doolittle 1997). These chaperones have the molecular function of assisting in the synthesis, folding, transport, and degradation of intracellular proteins (Srivastava 2002). High levels of expression of these proteins under conditions of cellular stress, as occurs during inflammation and immune responses against major foreign HSPs during bacterial infections, are currently being considered as a potent antigen in the development of dangerous diseases, such as autoimmune arthritis (Lopes-Silva et al. 2009). Mycobacterial HSP65 protein has been shown to provide protection against laboratory-induced arthritis stimulated by Streptococcus sp. cell-wall extracts (van Eden Broek 1989), Freund’s complete adjuvant (Billingham et al. 1990), pristano (Thompson et al. 1998), and collagen (Kingston et al. 1996). This protective effect has been related to the presence of certain peptides in the protein, which activate T cells and suppress antibodies that inactivate self-reactive T lymphocytes (Thorns and Morris 1985; Shoenfeld et al. 1986; Prescott et al. 2000).

The gene for HSP65 of Mycobacterium leprae has been shown to have immunomodulatory effects in various diseases, including the inhibition of the development of different types of tumor cells (Lukcas et al. 1993; Michaluart et al. 2008), the development of prophylactics against tuberculosis (Jones et al. 1993; Rosada et al. 2008; Souza et al. 2008; Zárate-Bladés et al. 2009) in systemic fungal infection (Ribeiro et al. 2009) and, potentially, the modulation of arthritis, diabetes and multiple sclerosis (van Eden et al. 1985; Life et al. 1993; Matzinger 1994; Chen et al. 1999; Quintana et al. 2003; Santos-Júnior et al. 2007). In arthritic rats, the purified HSP65 protein co-administered orally with soybean trypsin inhibitor at low doses (30 mg) has been shown to play an important role in the modulation of arthritis induced by adjuvant (AIA). This effect is related to the activation of regulatory T cells specific to the protein (Haque et al. 1996; Cobelens et al. 2000; Ulmansky 2002).

In the study reported here, we show the effectiveness of oral therapy using transgenic tobacco plants expressing the M. leprae HSP65 protein in the experimental model of AIA rats.

Materials and methods

Vector construction

The pcDNA3.hsp65 plasmid, which contains a 3.3-kb fragment that includes the M. lepraehsp65 gene, was kindly donated by Dr. Celio L. Silva. It was amplified by PCR using the oligonucleotides TUBCF (forward: 5′-GGAATTCCATGGCCAAGACAATTGCC-3′) and TUBCE (reverse: 5′-CGGGATCCCGTCAGAAGTCCATACCACCC-3′) under PCR conditions of 95°C for 10 min, followed by 35 cycles at 95°C for 1 min, 60°C for 50 s, and 72°C for 2 min, and a final extension at 72°C for 7 min. The PCR product was cloned into the multiple cloning site of the pGEMT-easy vector (Promega, Madison, WI). Digestion of this plasmid with EcoRI and BamHI released a 1,638-bp fragment which was then subcloned into the pUCpSS vector (Burlington, Canada). The pUCpSShsp65 plasmid was then digested with HindIII to release the cassette of 3,582 bp that was subcloned into the pCAMBIA 2301 plasmid, which is a plant expression vector. This final plasmid was named pChsp65.

Transformation and regeneration of tobacco plants

The pChsp65 plasmid was introduced into Agrobacterium tumefaciens LBA4404 by electroporation (Hoeckema et al. 1983; Mattanovich et al.1989) using a Gene Pulser electroporator (Bio-Rad, Hercules CA); transformation was confirmed by PCR amplification of the hsp6 gene. Transgenic plants were obtained by infecting Nicotiana tabacum TB1 leaf discs with recombinant A. tumefaciens-hsp65, following the protocol described by Baumann et al. (1987). The regeneration and selection of transgenic plants was carried out on solid MS medium (Murashige and Skoog 1962). A. tumefasciens cells were eliminated after transformation by culturing the transformed tissue for one round in medium containing kanamycin (100 mg/l) and claforan (500 mg/l). Regenerated transformed calli were subcultured in a medium containing claforan (500 mg/l) and kanamycin (200 mg/l), and regenerated shoots were subcultured in phytohormone-free MS medium. Rooted plants were transferred to a peat moss–agrolite (1:1) substrate and maintained under greenhouse conditions.

Histochemical analysis of tobacco plants

Fresh leaves from transgenic tobacco, wild-type (WT) tobacco, and tobacco-pCAMBIA 2301 (vector only) plants were submerged in β-glucuronidase (GUS) reaction buffer [0.1 M NaHPO4, pH 7, 1 mM K3Fe (CN)6, 1 mM K4Fe(CN)6, 10 mM EDTA, 0.1% Triton X-100 N-laurylsarcosil, 0.5 mg/ml−1 5-bromo-4-chloro-3-indolyl-beta-D-glucuronic acid cyclohexylammonium (Sigma, St. Louis, MO)] for 24 h at 37°C. Following incubation, the chlorophyll of the explants was removed with absolute ethanol by placing the samples in the organic solvent for 2 h (Jones and Sutton 1997). The characteristic blue staining indicating expression of the gus gene was determined by optical microscopy.

Nucleic acids extraction

For the extraction of DNA from transgenic plants, fresh leaves (100 mg) were frozen in liquid nitrogen and finely pulverized with a previously sterilized mortar. Genomic DNA was extracted following the protocol of Doyle and Doyle (1987). For RNA extraction, 1 g of fresh leaf tissue was finely pulverized in liquid nitrogen. RNA extraction was performed according to the protocol of López-Gómez and Gómez-Lim (1992).

Reverse transcription-PCR

A 3-μg aliquot of total RNA was used for reverse transcription (RT) in a 20-μg reaction volume using SuperScript II Reverse Transcriptase (Invitrogen, Carlsbad, UK). A 5-μl volume of this RT mixture was then used for the PCR analysis using primers TUBCF and TUBCE.

Western blot analysis and quantification of the M. leprae HSP65 protein in transgenic tobacco plants

Protein extracts were obtained from 100 mg of fresh pulverized leaves from both WT and transgenic plants. Samples were resuspended in 300 μl extraction buffer [1× phosphate buffered saline (PBS), 0.5% Tween 20, 10 mM DTT, 50 mM Tris-HCl (pH 7.4), 0.5 sucrose, 10 μg/ml phenylmethylsulphonyl fluoride, 5 μg/ml phosforamidon], shaken for 1 min, and then centrifuged at 14,000 rpm at 4°C for 5 min. The total protein concentration was determined by the method of Bradford (1976), and HSP65 protein quantification was carried out by densitometric analysis using the program RFLPScan V2.1 (Scanalytics, Fairfax, VA). HSP65 immunodetection was performed by electrophoresing 30 μg of protein extract from each of the plant lines, namely, WT (tobacco), tob-pCAM (transgenic tobacco vector only), and tob-HSP65 (transgenic tobacco expressing HSP65 protein), and 30 μg of extract from Escherichia coli (MH98ES) as a positive control in a polyacrylamide gel (12%). The proteins were then transferred to a nitrocellulose membrane (Bio-Rad, Hercules CA); the membrane was blocked with 5% nonfat milk in PBS at 37°C for 1 h and then incubated for 30 min with a 1: 2000 dilution of monoclonal antibody IIIE9 (epitope-specific, unique for M. leprae HSP65; Gillis and Buchanan 1982) in PBS–0.5% Tween (PBST) containing 1% nonfat milk at room temperature. The membrane was washed three times (5 min each time) with PBST. The rabbit anti-mouse immunoglobulin–peroxidase conjugate (Sigma) was diluted 1: 2000 in 1% nonfat milk–PBST. The membrane was incubated with the antibody at room temperature for 30 min, then washed four times with PBST. The washed membrane was incubated with AEC/DEF (3-amino-9-ethylcarbazole/N,N dimethylformamide) substrate to develop the protein bands. The molecular weights of the bands were estimated on the basis of a protein weight marker (Sigma).

Induction and clinical evaluation of AIA

Male Lewis rats, 6–8 weeks old and weighing 180–190 g (Harlan Mexico S. A. de CV, Mexico), were used. To induce arthritis, we injected the animals intradermally at the base of the tail with 200 μl of complete Freund adjuvant (2 mg M. tuberculosis strain H37Ra/rat; Sigma). Development of AIA was recorded at the start of the study and on days 1, 10, and 30. The body weight of the animals was determined, and the inflammation at the joint level in the hind feed of each animal was measured with a Mitutoyo micrometer (Mitutoyo Corp, Kawasaki, Japan).

Oral administration of the HSP65 protein expressed in transgenic tobacco

After the tenth day of AIA, a time at which the rats showed weight loss and joint inflammation, the rats were subjected to a 16-h fasting period. In the study, six groups of five AIA rats each were used and fed with leaf tissue equivalent to 10 μg of HSP65 protein. The same quantity of tobacco WT and tob-pCAMB leaf tissue were supplied to the control rats. Another group of animals was treated with 30 μg of purified recombinant M. leprae HSP65 protein (E. coli, pUC8-hsp65; kindly provided by R. Butler, NIMR, London, UK) co-inoculated with a soybean trypsin inhibitor (rHSP65-STI). Control groups of arthritic rats and healthy rats without treatment were included.

Statistical analysis

Group differences in cumulative arthritis scores of weight and inflammation in disease incidence were analyzed using Student’s t test. P values less than 0.05 were considered to be statistically significant.

Results

pChsp65 vector construction

PCR amplification of HSP65 of M. leprae with primers designed to correspond to the complete gene yielded a 1,644-bp product; this fragment was subcloned, and the pChsp65 plasmid was generated. This plasmid has two reporter genes, the kanamycin resistance gene and the gene encoding the GUS enzyme. The hsp65 gene was cloned in a cassette with a double cauliflower mosaic virus (CaMV) 35S promoter and a CaMV 35S terminator region in the plasmid pCAMBIA 2301 [Electronic Supplementary Material (ESM) Fig. 1S].

Histochemical analysis of transgenic plants

Plasmid pChsp65 was used to produce transformed tobacco plants; incorporation of the plasmid was confirmed by the survival of plants (>85%) for over 1 month in kanamycin medium. None of these plants showed apparent phenotypical changes. Published reports indicate that the success rate of obtaining transformed plants of a particular species lies in the degree of survival that species shows when exposed to an antibiotic, without the appearance of pleiotropic effects that would affect the normal development of the plant (Nölke et al. 2005). Transformation was further confirmed with GUS assay: of the 40 lines chosen to be assayed, 42.5% (lines 3, 6, 7, 8, 9, 10, 11, 12, 17, 20, 23, 31, 32, 33, 34, 38, 40) presented different degrees of staining, a parameter indicative of enzymatic activity (ESM Fig. 2S A). The organs that depicted the highest degree of staining were the trichomes and veins (ESM Fig. 2S B, C, D). No staining was observed in any of the wild-type (ESM Fig. 2S A; lines 13, 14, 15, 16) and tobacco-pCAMBIA 2301 (vector only) plants analyzed (ESM Fig. 2S A, lines 3, 6, 10). These results confirm the integration and stable expression of the gus gene in the genomic DNA of the transformed tobacco plants. The activity of the GUS enzyme in roots and leaves of some plant species transformed with the plasmid pCAMBIA 2301 has been reported (McIntosh et al. 2004).

PCR confirmation of the presence of hsp65 in transgenic tobacco plants

We chose transgenic tobacco lines 7, 9, 20, 31, and 34 as test material to confirm the incorporation of the hsp65 gene into the genome of the tobacco plants. Using genomic DNA extracted from transgenic tobacco and control plants, we PCR-amplified the M. leprae hsp65 gene using the oligonucleotides TUBCE and TUBCF. Negative controls included the amplification of genomic DNA of WT tobacco plants and tobacco plants transformed with pCAMBIA 2301. The amplified products were analyzed electrophoretically, and a 1,644-bp fragment corresponding to the hsp65 gene was found only in transgenic plants. No comparable fragment was amplified in the control plants (ESM Fig. 3S A). Amplification of the hsp65 gene provided evidence of its incorporation into the genomic DNA of the different lines of transformed plants. Of the lines chosen for the GUS analysis, only lines 7, 9, 20, 31, and 34 showed the PCR product; those lines transformed with the plasmid pCAMBIA 2301 and the WT plants did not present this 1,644-bp product (ESM Fig. 3S A).

Confirmation of hsp65 gene expression in transgenic tobacco plants was carried out by RT-PCR analysis. cDNA synthesis was performed using 3 μg of total RNA. Using this cDNA and the hsp65 primers in a PCR reaction resulted in the amplification of a 1,644-bp fragment in the putative transgenic lines; this product was not obtained in the WT lines or in those transformed with the plasmid alone (ESM Fig. 3S B). These results demonstrated that it was indeed feasible to express the hsp65 gene of mycobacterial origin in the plant genome.

Expression of the HSP65 protein

To determine which transgenic line produced the highest levels of HSP65 protein, we performed a western blot assay using monoclonal antibody IIIE9 (which only recognizes M. leprae HSP65). The results showed that the different lines expressed different levels of protein, with line 34 showing the highest level of expression (0.19% of the total soluble protein) (Fig. 1A, B).

Fig. 1
figure 1

Analysis of heat shock protein 65 (HSP65) in tobacco plants. A Western blot analysis of transformed tobacco using the first monoclonal antibody IIIE9 and a second anti-rabbit antibody conjugated with the peroxidase enzyme. Presence of a 65-KDa band in transformed tobacco lines containing the hsp65 gene (tob-hsp65; lines 7, 9, 20, 31, 34) and in those containing recombinant HSP65 (rHSP65; E. coli MH9ES), but not in wild-type tobacco (WT) plants and plants with tobacco-pCAMBIA 2301 (tob-pCAM). B Densitometric analysis of the expression of HSP65 protein in transgenic tobacco lines containing the hsp65 gene (tob-hsp65): 7 (0.09%), 9 (0.098%), 20 (0.13%), 31 (0.12%), 34 (0.19%)

Full size image

Oral administration of the HSP65 protein expressed in tobacco to arthritic rats

Throughout the 30 days of observation, the group of healthy rats (without treatment) showed significant weight gain (t0d = 187 ± 6.9, t10d = 223.6 ± 4.7, t30d = 263.66 ± 1.52 g), but not a gain in joint measurements (t0d = 6.02 ± 0.05, t10d = 6.24, t30d = 6.2 ± 0.19 mm) (Fig. 2A, B). In contrast, rats treated with the adjuvant lost significant weight (t0d = 190 ± 3.03, t10d = 164.6 ± 9.8 g, t30d = 159 ± 1.15 g) (Fig. 2A), and the size of their joints increased (t0d = 5.78 ± 0.05, t10d = 7.28 ± 0.05, t30d = 10.41 ± 0.10 mm) (Fig. 2B).

Fig. 2
figure 2

Effect of oral administration of mycobacterial HSP65 protein expressed in tobacco plants after the development of adjuvant induction arthritis (AIA) on weight (A) and joint inflammation (B). On day 0, rats were immunized with 2 mg of Mycobacterium tuberculosis (strain H37Ra)/rat in Freund’s complete to induce AIA. On day 10, treatment was started with oral administration of transgenic tobacco plants containing the hsp65 gene (tob-hsp65), tobacco-pCAMBIA 2301 plants (tob-pCAM), wild-type tobacco plants (WT), and plants containing the recombinant Hsp65 protein (rHSP65-STI). Healthy and diseased control rats received no treatment, and on day 30, rats were killed by decapitation. The rats were examined daily. Student’s t test results are expressed as the mean ± standard deviation. *P < 0.05 was considered to be significant, **P < 0.01 when compared to the control groups

Full size image

Animals fed with WT tobacco and tob-pCAMBIA 2301 plant tissue showed a recovery of weight and a reduction of joint inflammation at the end of treatment (WT tobacco: t0d = 189.2 ± 3.6, t10d = 161.8 ± 2.70, t30d = 190.66 ± 2.50 g; t0d = 5.84 ± 0.1, t10d = 7.2 ± 0.18, t30d = 8.2 ± 0.05 mm, respectively; tob-pCAMBIA 2301: t0d = 193 ± 5.17, t10d = 172.00 ± 8.89, t30d = 187.00 ± 2.08 g; tod = 5.5 ± 0.2, t10d = 7.5 ± 0.23, t30d = 8.60 ± 0.25 mm, respectively) (Fig. 2A, B). Weight recovery and joint deinflammation were significantly higher than that of the non-treated arthritic rat group.

In contrast to the above-mentioned groups, the group of rats that received tobacco-hsp65 leaves as food markedly increased their body weight (t0d = 189.2 ± 4, t 10d = 169.6 ± 2.8, t30d = 232.33 ± 1.7 g), and the size of their joints decreased significantly (t0d = 5.52 ± 0.1, t10d = 7.76 ± 0.19, t30d = 6.3 ± 0.05 mm) (Fig. 2A, B). Weight recovery and reduction in joint inflammation was significantly higher with this treatment than with the other treatments, with the animals almost reaching the joint values of the “healthy” individuals.

Animals treated with rHSP65-STI showed a similar behavior as those fed tobacco-hsp65 leaves, with a weight gain (t0d = 192.4 ± 2.8, t10d = 167.8 ± 12.81, t30d = 192.66 ± 6.11 g) and reduced inflammation (t0d = 5.7 ± 0.4, t10d = 7.5 ± 0.05, t30d = 7.0 ± 0.06 mm) (Fig. 2A, B). However, these values are significantly different from those in rats receiving tob-hsp65 leaves (see Fig. 2A, B) whose weight recovery and reduction in joint inflammation were relatively better.

Discussion

Our results demonstrate that tobacco plants infected with A. tumefaciens carrying the plant expression plasmid pChsp65 stained positive in the GUS histochemical assay. We detected the expression of the GUS protein in different tissues of the plants, principally in veins (vascular tissue) and trichomes. This expression is in agreement with that reported by Jefferson et al. (1987) using the GUS gene under control of the 35S promoter.

The mycobacterial hsp65 gene and the recombinant protein have been used as a potential therapeutic agent (gene therapy) to treat animals with autoimmune diseases, such as AIA in rats and diabetes in non-obese mice (Elias et al. 1990; Cobelens et al. 2000 Nomaguchi et al. 2002; Santos-Júnior et al. 2007). Some studies report that the oral administration of the HSP65 protein to arthritic rats induces an important weight recovery and reduction of joint inflammation together with a reduction in the cartilage- and bone-destroying process (Billingham et al. 1990; Cobelens et al. 2000). Our results with respect to the oral administration of transgenic plant tissues that express the HSP65 protein are in agreement with those of Cobelens et al. (2000), who reported the oral administration of the recombinant protein (expressed in E. coli) in conjunction with a soybean trypsin protein to rats with AIA. We founded a significant effect on recovery weight (Fig. 2A) and, in the case of joint inflammation, the results were also similar but obtained higher values since changes were observed from the first administration (Fig. 2B). One important point is that we fed the rats plant tissue without trypsin inhibitor and in quantities of 10 μg. One explanation for these interesting results is the possibility of a synergistic effect between the HSP65 protein and some compounds present in tobacco plants. We also observed a deinflammatory effect in those rats fed with WT tobacco tissue or tobacco-pCAMBIA 2301 tissue; this could be due to the effect of nicotine and/or scopoletin, both of which have been shown to have anti-inflammatory properties (Kalra et al. 2004; Kim et al. 2004). The oral administration of transgenic tobacco HSP65 in five doses to arthritic rats was enough to counteract the effects of the disease symptoms. This effect could ultimately be the consequence of a modulation of the immune system (van Eden Broek 1989; Cobelens et al. 2000; Kalra et al. 2004; Kim et al. 2004; van Eden et al. 2005).

Our data suggest that M. leprae HSP65 protein produced by plant tissues was successful in mitigating the effects of AIA, perhaps with the assistance of synergistic effects with plant metabolites. These results confirm the heterologous expression of the mycobacterial HSP65 protein in tobacco plants and its potential effect when orally administered for the treatment of arthritis. To the best of our knowledge, this is the first report of a synergistic effect in transgenic plants.