Using targeted toxins is a promising approach for the therapy of cancer and autoimmune diseases, as well as other disorders.1 The high mobility group chromosomal protein N2 (HMGN2) is one of the most abundant and well-characterized classes of nonhistone nuclear proteins, which seems to function as architectural elements in chromatin.2 Recently our group isolated an antimicrobial polypeptide from human LAK cells and cervical mucus, which was characterized to be the HMGN2.3–7 We also found that the transmembrane α-helical structure located in the 17–47 residues, which has been found to be the DNA binding domain of HMGN2, was essential for its antibacterial activity.8 Porkka et al9 recently reported that HMGN2 α-helical fragment could specifically bind to tumor cells in vitro and in vivo. HMGN2 α-helical domain could carry a payload (phage, fluorescein) into the nuclei of tumor cells and endothelium of tumor-associated blood vessels. Thus, the HMGN2 α-helical fragment could presumably be used to target cytotoxic drugs for cancer therapy. In this study, we created a recombinant chimeric toxin composed of HMGN2 α-helical domain and pseudomonas exotoxin domain III and examined its anti-cancer activity against Hela cells in vitro and in vivo.
Human HeLa cell lines were kept in our lab and cultured in 10% bovine serum in DMEM containing L-glutamine, penicillin, and streptomycin. The human mononuclear leukocytes were isolated from human blood and cultured in RPMI 1640 containing penicillin and streptomycin in the presence of 10% bovine serum, and stimulated with IL-2 and phytohemagglutinin (PHA). Human cervical tissues were collected from uterus of uteroectomy women. Isolation and culture methods were described elsewhere. The cell viability was confered by 0.1% trypan blue.
Construction of plasmids for expression of recombinant HMGN2-PE III
Total RNA was isolated with Trizol Reagent (Gibco BRL, USA) from IL-2-activated human mononuclear leukocytes. HMGN2 α-helical domain cDNA was amplified by RT-PCR. Generation of DNAs encoding HMGN2 α-helical domain and pseudomonas exotoxin domain III was performed by PCR amplification. Primers introduced an EcoRI and Hind III restriction site (underlined) were designed as follows: P1 (5′GCGAATTCATGAAGGACGAACCACA3′) and P2 (5′TGATCCACCACCTGACTTCTTTGCAGGGGC3′) for HMGN2 α-helical domain amplification; P3 (5′TCAGGTGGTGGATCAGGCCCGGCGGACAGC3′) and P4 (5′GGCAAGCTTAGAGCTCGTCTTTCGG3′) for pseudomonas exotoxin domain III amplification. Around 100 bp HMGN2 α fragment and 700 bp PEIII fragment were obtained.
P1 and P4 were used to link HMGN2 α-helical domain and pseudomonas exotoxin domain III by SOEing PCR, and an 800 bp HMGN2α-PEIII chimera was obtained. After digestion with EcoRI and HindIII, the PCR products were inserted into pET-32a (+) (Novagen, USA). DNA sequencing of the recombinant prokaryotic expression vectors pET-32 a (+)-HMGN2α-PEIII was carried out to confirm the inserted sequence.
Production of the chimeric protein
Plasmids encoding HMGN2 α-PEIII were amplified and purified as described previously.5 The transformed E. coli BL21 (λDE3) carrying pET-32 a (+)-HMGN2α-PEIII was cultured in LB medium for 12 hours in the presence of 0.1 mmol/L IPTG to induce protein expression. The induced cultures were washed with PBS and the cell lysates were obtained by freezing/thawing in the presence of lysozyme. After centrifugation, the supernatants were saved. The fusion proteins were purified through HisTrap Chelating HP columns (Pharmacia, USA) then cleaved by thrombin digestion. The chimeric protein HMGN2α-PEIII was obtained by RP-HPLC purification. Protein concentration was determined by bicinchoninic acid BCA Protein Assay Kit (Pierce, USA) and the purity was confirmed by Tricine-SDS-PAGE and Western blot with a primary antibodymouse anti-His6 monoclonal antibody (Roche, UK).
The cytotoxicity of HMGN2α-PEIII to HeLa cells and human cervical epithelial cells was assessed by MTT, at a concentration of 2×104 cells/well, plated in 96-well plates and incubated overnight. The chimeric protein was diluted in PBS/0.2% BSA to desired concentrations and added to target cells in triplicate. The cells were incubated for 48 hours at 37°C before adding MTT per well and further incubated for 4 hours at 37°C. After removing the media, DMSO (200 μl/well) was added to dissolve the cell pellet. The absorption was measured at 590 nm on BIO-RAD Model 550 microplate reader.
Fluorescein isothiocyanate (FITC)-labeling of the chimeric proteins and determination of cellular internalization FITC was purchased from Calbiochem-Novabiochem Corp. (San Diego, USA). Two milligram of proteins was dissolved in 1 ml of 0.02 mol/L Na2CO3-NaHCO3 buffer, pH 9.1, containing 0.02 mol/L NaCl. To the protein solution, 500 ml of 1% (w/v) FITC dissolved in acetone was added, and the reaction mixture was kept at 25°C for 5 hours. The reaction was terminated by dialysis against water and further purified by C18 reverse-phase HPLC. The purified FITC-labeled proteins were dried under vacuum condition and resuspended in water. Their concentrations were determined by BCA method and the labeling rate was confirmed by measuring the absorption of λ280 and λ495 nm.
To determine the internalization of the chimeric protein in HeLa cells and human cervical epithelial cells, the cell suspensions were plated at a concentration of 2 × 105cells/well in 6-well plates and incubated at 37°C overnight and then 500 μg/ml FITC-labeled proteins (50 μl/well) were added at final concentration of 12.5 μg/ml and incubated for 24 hours at 37°C. After washing 5 times with PBS, the samples were fixed with paraformaldehyde, and then observed under fluorescence microscope.
DNA binding assay
The genomic DNAs were extracted from HeLa cells and human cervical epithelial cells using routine method. Gel-retardation experiments were performed by mixing 100 ng of the genomic DNA with increasing amount of the chimeric protein in 0.5 ml of binding buffer.11 The reaction mixtures were incubated at room temperature for 3 hours. Subsequently, 100 μl of loading buffer was added and an aliquot of 12 μl was applied to a 1% agarose gel electrophoresis in 0.5 L Tris borate-EDTA buffer.
In vivoanti-tumor assay
The in vivo anti-tumor activity of the chimeric protein was determined in female Balb/c nude mice (grade III, certificate No. 9, about 20 g, 2-month-old) bearing human cancer cells. HeLa cells (2×106) were inoculated subepidermally in the left oxter of the nude mice. By day 14, the tumors (about 0.5 cm3 in size) developed in all the tumor-implanted animals. The therapy was started on day 14. In the treatment group (6 animals), the chimeric protein (12 mg/kg body weigh) diluted in 0.2 ml was given by intraperitoneal injection for 3 times at one day interval. In the control group (4 animals), N.S. was given in the same way. Tumors were measured with a caliper every 2 or 3 days. The animals were sacrificed in 24 hours later after the last treatment. Tumor was taken and weighted. The rate of tumor inhibition (%)=(tumor tissue weight of control group-tumor tissue weight of the chimeric protein, treated group)/tumor tissue weight of control group × 100%. At the same time, heart, liver, lung, spleen, kidney, uterus were also taken, fixed with 10% formalin. The tissue sections were stained with HE for histological examination.
Tumor sizes are expressed as mean ± standard deviation (SD). The differences between treatment group and the control were determined by t test. P<0.05 was considered to be significant statistically.
Production and purification of the recombinant chimeric protein
The recombinant fusion protein was produced in E. coli DE3 transformed by the recombinant plasmids and purified by affinity chromatography to around 95% purity determined by SDS-PAGE. The purified 50 kD recombinant fusion protein was cleaved by thrombindigestion, and the 40 kD recombinant chimeric protein was obtained via RP-HPLC purification (Figure 1 left, right). The cell lysates of the recombinant-transformed and IPTG-induced bacteria and native E. coli BL21, and the purified fusion protein were subjected to SDS-PAGE. The blots were probed with mouse anti-His6 monoclonal antibody. Specific signals were detected on the cell lysate of the recombinant-transformed bacteria and the purified fusion protein. In contrast, no signal was seen on the cell lysate of native E. coli (Figure 1 middle).
Cytotoxicity of the chimeric protein
To determine whether the chimeric protein would be cytotoxic, HeLa cells or human cervical epithelial cells were incubated with the chimeric protein for 48 hours, followed by MTT analysis. Figure 2 showed that the chimeric protein was cytotoxic to both of HeLa cells and human cervical epithelial cells in a concentrationdependent manner. However, its cytotoxicity to HeLa cells was much higher than to human cervical epithelial cells. To HeLa cells, the LC50 was 12.86 μg/ml, and 95% confidence interval was 6.63–24.94 μg/ml calculated by Bliss analysis, and to human cervical epithelial cells, the LC50 was 57.74 μg/ml, and 95% confidence interval was 12.05–276.71 μg/ml. We also compared with poclitaxel, an anti-cancer drug in this assay system, and found that 60 ng/ml poclitaxel could kill 55.5% HeLa cells and 23.7% normal human cervical epithelium cells. This result suggested that the cytotoxicity of the chimeric protein had some specificity to HeLa cells.
Internalization of the chimeric protein
To observe the internalization of the chimeric protein in HeLa cells, the protein was labeled with FITC fluorescence. HeLa cells and human cervical epithelial cells were incubated with the FITC-labeling chimeric protein for 24 hours. The fluorescence microgram showed that the FITC fluorescence was seen in some HeLa cells. The rate of cell fluorescence positive was about 31%. In contrast, almost no fluorescence could be seen in human cervical epithelial cells. This preliminary result suggested that the chimeric protein could exhibit some specificity for the recognition of cancer cells.
Binding of the chimeric protein to genomic DNA
The gel electrophoretic mobility shift indicated that the chiemric protein suppressed the migration of the genomic DNA of HeLa cells at 1:10 weight ratio of DNA to the protein. However, even at 1:100 DNA to protein ratio, the suppression was not observed in the human cervical epithelial cells in this assay condition. This preliminary result suggested that the chimeric protein may preferentially bind to the genomic DNA of HeLa cancer cells.
Anti-tumor effect on nude mice bearing HeLa tumors
To determine the in vivo anti-tumor potency of the chimeric protein, the nude mice bearing HeLa tumors were treated with 12 mg/kg the protein on day 14, 16, and 18 in all three times. Figure 3 shows the tumor sizes on day 19. The tumor weight of the chemeric protein-treated group was much smaller than that of N.S.-treated control group ((0.1093±0.0424)g vs (0.4446±0.2033)g, P<0.01). Based on the weight, the tumor inhibition rate of the chimeric protein was 75.4%.
In the histological examination, large area of necrosis and some apoptosis could be seen in the tumors of the chimeric protein-treated group. In the heart, liver, lung, spleen, kidney and uterus, there was no any obvious abnormality in the chimeric protein treatment mice compared with N.S. and control group.
Drug's targeting is important for cancer therapy. Recently, followed by the development of genetic engineering, researchers have been playing more attention to the targeting therapy of recombinant immunotoxin. The recombinant immunotoxin can specifically target and bind cancer cells, and accumulate in the endemic area, leading to improvement of its cure efficacy and depression of its toxicity and damage to normal tissues. PE, an exotoxin derived from Pseudomonas areuginosa, is a 66-kD molecule that kills animal cells by inhibition of protein synthesis.11 Crystallographic studies have identified three major domains in PE. Domain Ia enables PE to bind to the PE receptor, and domain II is involved in the processing and translocation of the toxin to the cytosol. Domain III contains the ADP-ribosylating activity that can catalyze the ADP-ribosylation of elongation factor-2. The function of domain Ib, a 4-kD fragment between domains II and III, is unknown, and deletion of most of its sequences can not result in the loss of PE activity. In addition, amino acids at the COOH-terminal residues of PE, REDLK, are required to route the toxin to an intracellular compartment. If the COOH terminus, REDLK sequence, is changed to the known endoplasmic reticulum retention sequence, KDEL, PE toxicity is increased. PE38 KDEL, in which DIa and the 365–388 aa in DIb are deleted, contains 38 kD PE fragment, and its COOH terminus is changed to KDEL sequence. PE38KDEL has been most frequently used as a toxin to construct recombinant immunotoxin recently. A lot of researches showed that the recombinant immunotoxin containing PE38KDEL can lead target cell to necrosis and facilitate apoptosis.12 Generally, recombinant immunotoxin is often made from the linkage of Fab of monocolonal antibody with PE domains II and III. Fab recognizes cancer cells as does PE domain Ia, but former is specific and the later is non-specific in the recognition. Domains II and III function its intracellular translocation and cytotoxicity respectively.13
In the current study, we constructed a recombinant chimeric toxin composed of HMGN2 α-helical domain and PE38KDEL domain III. Experiments indicated that this chimeric protein exhibited its in vitro cytotoxicity and genomic DNA binding activity much stronger to HeLa cancer cells than to normal epithelial cells. In FITC fluorescein-labeling experiments the preliminary result suggested that HMGN2 α helical sequence probably made this chimeric toxin penetrate and accumulate in HeLa cancer cells. The in vivo experiments also showed that systemic administration of this chimeric protein significantly inhibited the growth of xenograft HeLa tumor in nude mice. Remarkable necrosis and apoptosis of tumors were seen in treated mice, but no histological changes were observed in the mouse organs, such as liver, kidney, heart, lung, etc.
Based on the preliminary experiment data and the known knowledge of the biological functions of PE domain III and HMGN2 α-helical domain, the anti-HeLa cell mechanisms of this chimeric toxin might presumably be related to the combination of inactivation of elongation factor-2 resulting in inhibition of protein synthesis, interfering DNA transcription, and facilitating apoptosis in cancer cells.
The HMGN2 α-helical sequence is only 30 aa long, which would reduce the molecular mass of chimeric toxin. Our preliminary study and other study9 suggested that HMGN2 α-helical domain might exhibit both intracellular translocation function and cancer targeting activity. HMGN2 α-helical sequence might be considered as an novel toxin-carrier in the study of cancer targeting therapy.
1. Pasetto LM, Bortolami A, Falci C, Sinigaglia G, Monfardini S. Recent progress in target therapy in colorectal cancer. Anticancer Res 2006; 26:3973–3981.
2. Trieschmann L, Postnikov YV, Rickers A, Bustin M. Modular structure of chromosomal proteins HMG-14 and HMG-17: definition of a transcriptional enhancement domain distinct from the nucleosomal binding domain. Mol Cell Biol 1995; 15: 6663–6671.
3. Yang HR, Huang N, Wu Q, Wang B. Isolation and characterization of an antibacterial polypeptide from human LAK cells. Chin J Microbiol Immunol (Chin) 2003; 23: 241–246.
4. Yang H, Pan X, Huang N, Wu Q, Feng Y, Wang B. Isolation and purification of an antibacterial polypeptide from human cervical mucus. J Sichuan Univ (Med Sci Edit) (Chin) 2003; 34: 214–216.
5. Xiong WB, Feng Y, Wang GX, Huang N, Wu Q, Li X, et al. Application of HMGN2-tag constructs to analysis of HMGN2 distribution in HeLa cells. J Biomed Eng 2005; 22:1015–1019.
6. Feng Y, Xiong W, Wang G. Identification of HMGN2 as an effector molecule of human LAK cells. Prog Biochem Biophys 2005; 32: 568–575.
7. Feng Y, Huang N, Wu Q, Wang B. HMGN2: a novel antimicrobial effector molecule of human mononuclear leukocytes? J Leukoc Biol 2005; 78: 1136–1141.
8. Feng Y, Huang N, Wu Q, Bao L, Wang BY. Alpha-helical domain
is essential for antimicrobial activity of high mobility group nucleosomal binding domain 2 (HMGN2). Acta Pharmacol Sin 2005; 26: 1087–1092.
9. Porkka K, Laakkonen P, Hoffman JA, Bernasconi M, Ruoslahti E. A fragment of the HMGN2 protein homes to the nuclei of tumor cells and tumor endothelial cells in vivo.
Proc Natl Acad Sci USA 2002; 99:7444–7449.
10. Gorodeski GI, Romero MF, Hopfer U, Rorke E, Utian WH, Eckert RL. Human uterine cervical epithelial cells grown on permeable support a new model for the study of differentiation and transepithelial transport. Differentiation 1994; 56: 107–118.
11. Park CB, Kim HS, Kim SC. Mechanism of action of the antimicrobial peptide buforin II: buforin II kills microorganisms by penetrating the cell membrane and inhibiting cellular functions. Biochem Biophys Res Commun 1998; 244: 253–257.
12. Sampson JH, Akabani G, Archer GE, Bigner DD, Berger MS, Friedman AH, et al. Progress report of a Phase I study of the intracerebral microinfusion of a recombinant chimeric protein composed of transforming growth factor (TGF)-alpha and a mutated form of the Pseudomonas exotoxin
termed PE-38 (TP-38) for the treatment of malignant brain tumors. J Neurooncol 2003; 65: 27–35.
13. Onda M, Wang QC, Guo HF, Cheung NK, Pastan I. In vitro
and in vivo
cytotoxic activities of recombinant immunotoxin 8H9(Fv)-PE38 against breast cancer, osteosarcoma, and neuroblastoma. Cancer Res 2004; 64: 1419–1424.