A Molecular Carrier to Transport and Deliver Cisplatin into Endometrial Cancer Cells

Authors


Corresponding author: Aldo Mancini, aldo_mancini@tiscali.it

Abstract

The leader peptide of a recombinant manganese superoxide dismutase (rMnSOD-Lp) acts as a molecular carrier. Clonogenic tests on normal (MRC-5) and endometrial adenocarcinoma cells (HTB-112) were carried out in the presence of rMnSOD-Lp, cisplatin alone (CC) or cisplatin conjugated to the rMnSOD-Lp (rMnSOD-Lp-CC). The platinum delivered into the cells was measured by atomic spectrophotometric absorbance. The treatments on tumor and normal cells were finally evaluated by LM and TM microscopy. Tumor cell death in the case of 0.5 μm cisplatin on its own was minimal, while in the presence of 0.5 μm rMnSOD-Lp-CC, no tumor cells survived. Atomic absorbance analysis showed that rMnSOD-Lp-CC delivered approximately four times more cisplatin into HTB-112 cells than the amount delivered using cisplatin alone. By LM observation, the cells treated with rMnSOD-Lp-CC showed signs of nuclear and cytoplasmic fragmentation, that is, apoptosis induced by the treatment. The therapeutic effect of rMnSOD-Lp-CC on endometrial cancer cells was significant, while on the normal cells it showed only a minimal toxicity. We believe that rMnSOD-Lp deserves to be considered as a molecular carrier to deliver cisplatin directly into tumor cells, thus transforming its antireplicative activity into a specific and selective antitumor agent.

Endometrial cancer (EC) is the sixth most diagnosed cancer in women (1). Surgical treatment represents the milestone of therapy for patients affected by endometrial carcinoma. Total hysterectomy with bilateral salpingo-oophorectomy and peritoneal washing either as a laparotomic or laparoscopic procedure is the appropriate surgical solution when the disease is confined to the uterus (2). The role of lymphadenectomy in case of myometrial invasion is under investigation (3,4).

Survival is directly related to the surgical stage, and the role of adjuvant radiotherapy in EC remains controversial. Several studies have demonstrated that external pelvic radiotherapy reduces isolated local recurrence (5,6), while adjuvant chemotherapy is recommended for patients with advanced disease; its combination with radiotherapy is still under debate (7).

Platinum (Pt) compounds, anthracyclines, and more recently taxanes have been used alone or in combination regimens, achieving response rates exceeding 50% and resulting in more than 1-year survival in randomized trials (8–14).Toxicity is a crucial point in this process, especially when associated with advanced age and multiple comorbidities.

To reduce the side effects, we have explored a new molecular approach.

Recently, we have isolated an isoform of manganese superoxide dismutases from a human liposarcoma cell line. The protein has also been obtained in recombinant form, known as rMnSOD. This enzyme catalizes the dismutation of O2– free radicals in H2O2, thereby preventing the accumulation of this activated oxygen species. H2O2 can be further converted into H2O and O2 by catalase and/or glutathione peroxidase. Once rMnSOD penetrates the cancer cells, it exerts its specific enzymatic activity consisting in the transformation of free radicals into hydrogen peroxide. It is noteworthy that catalase, the enzyme responsible for H2O2 detoxification into molecular oxygen, is usually present in amounts 10–50 times lower in many tumor cells than in their normal progenitors(15). Treatment with rMnSOD may therefore lead to an especially high accumulation of H2O2 in these neoplastic cells, resulting in their specific killing (16). Moreover, during the characterization of rMnSOD, we have demonstrated that the unusual characteristic of rMnSOD of penetrating tumor cells was attributed to its leader uncleaved peptide of rMnSOD that has the function of a molecular carrier (16). By considering the ability of this carrier to penetrate cells, we have recently shown that if it is cisplatin-conjugated, the molecular carrier enters the cultured cells and releases cisplatin into the cell cytoplasm, therefore greatly increasing the therapeutic index of cisplatin. With this system, we have transformed the antireplicative activity of cisplatin into a specific antitumor agent (17).

In the present study, we have evaluated the antitumor activity of the construct rMnSOD-Lp-CC on the human endometrial adenocarcinoma cell line HTB-112 and on human normal cells MRC-5. The rationale for the choice of our model was based on the fact that endometrial carcinoma patients respond well to cisplatin treatment (18).

Materials and Methods

Synthesis of rMnSOD-LP-CC

The peptide was synthesized step-wise in batch-mode on solid support with automatic synthesizer Syro (MultiSynTech GmbH, Witten, Germany- Rapp Polymere GmbH, Tuebingen, Germany), using Fmoc/tBu chemistry (19) starting from the C-terminal residue preloaded on PS-PHB resin with an average of 0.57 mmol/g substitution (Rapp Polymere). Synthesis on a 30-micromol scale proceeded through standard cycles of Fmoc deprotection [piperidine/DMF, 0.2:1(v/v)] and Fmoc-amino acid coupling steps [Fmoc/amino acid/HBTU/DIEA, 1:4:4:8] (20).

As Pt binder, diamino-ethyl-glycine was used, which, by virtue of the presence of two free amine groups, is able to combine platinum(II) ions as PtCl2 (21).

Amino acid derivatives

Fmoc-Gln(Trt)-OH, Fmoc-Arg(Pbf)-OH, Fmoc-Ser(tBu)-OH, Fmoc-Gly-OH, Fmoc-Leu-OH, Fmoc-Tyr(tBu)-OH, Fmoc-Ala-OH, Fmoc-Pro-OH, Fmoc-Thr(tBu)-OH, Fmoc-Cys(Trt)-OH, Fmoc-Val-OH, Fmoc-Met-OH.

Binder

N-Fmoc[N’-Fmoc-(2′aminoethyl)]glycine.

Cleavage

Side chain de-protection with concomitant cleavage of peptide from solid support was achieved by suspending the protected peptide-resin in a TFA/water mixture [95:5 (v/v)] for 3 h. The resin was removed by filtration under reduced pressure. Precipitation was achieved by collecting the filtrate on a bed of cold ether. After several ether washes, the crude compound was suspended in water (0.1% TFA, v/v) and lyophilized.

Purification

RP-HPLC purification was performed on Vydac C18 (22 × 250 mm); eluent A, 0.1%TFA in water; eluent B, 0.1%TFA in acetonitrile; gradient, from 5%B to 65%B in 20 min; flow, 20 mL/min.

Analysis

Purity was assessed by analytical RP-HPLC (Shimadzu): Vydac C18 column (4.6 × 150 mm); eluent A, 0.1%TFA in water; eluent B, 0.1%TFA in acetonitrile; gradient, from 5%B to 65%B in 20 min; flow, 1 mL/min; (R.t.), 17.7 min. The analytical HPLC chromatogram showed that the final product was obtained with hight purity grade (95.7%). Identity was assessed by mass spectrometry performed on a MALDI-T of spectrometer (Applied BioSystem), and the mass analysis confirmed the presence of desired compound: [MH+] = 2902 m/z (calcd = 2901).

Atomic absorbance analysis

Target cells: three samples of HTB-112 target cells used in the present study were treated for 3 h in the presence of 92 μm of cisplatin or 92 μm of rMnSOD-Lp-CC (containing an equivalent 11.1 μg of cisplatin) in culture in DMEM with 10% FCS (Gibco); the controls were obtained by maintaining the same cells in identical culture condition but in the presence of 92 μm of peptide alone in the absence of cisplatin. Target cells were collected, washed twice in phosphate buffer (PBS), and treated with 50 μL of 35% HNO3 for 16 h. Platinum content in the growth medium and in the cellular pellet after 3 h of incubation was determined by graphite furnace atomic absorption spectrophotometry (Analyst 800; Perkin-Elmer, Norwalk, CT, USA) using the following parameters: pretreatment temperature, 1300 °C; atomization temperature, 2200 °C, with 0.015 mg of Pt plus 0.01 mg of Mg(NO3)2 as matrix modifier. The measurements were performed using a graphite furnace equipped with Zeeman-effect background correction system (22). This analysis had been previously performed on the normal MRC-5 cells (17). Pyrolytic graphite-coated THGA tube (Perkin-Elmer) with an integrated Lvov-type platform was used for the metal determination. A Pt standard in 2.5% HNO3 (Spectrascan, Shimadzu, Kyoto, Japan) was used as a stock solution for the construction of the three-point calibration curve. Each measurement was carried out in triplicate.

Clonogenic tests on normal and tumor cells following rMnSOD-Lp-CC treatment

Confluent 75 cm2 flasks of tumor (HTB-112) or normal (MRC-5) cells were trypsinized, counted with a hemocytometer, and diluted in complete media to obtain 100 cells/mL. Two milliliters of cell suspension was plated in each well of a six-well tissue culture plate to obtain 200 cells per well. Normal and tumor cells were treated in the presence of rMnSOD-Lp-CC at final concentration ranging from 0.03 to 0.5 μm. Experiments were performed in triplicate, and clonogenic tests were performed on the following different cells groups:

• HTB-112 cells and MRC-5 cells cultured in their specific medium and considered as negative control.

• HTB-112 cells and MRC-5 cells treated in the presence of cisplatin alone.

• HTB-112 cells and MRC-5 cells treated in the presence of rMnSOD-Lp-CC.

Colonies were stained with crystal violet after 14 days, and those containing at least 30 cells were counted as surviving colonies. The plating efficiency and the survival fraction, for each cell line after each treatment, were calculated according to the method proposed by Franken et al. (23) and averaged approximately 80% for all cell lines. Survival was calculated in comparison with non-treated samples, using an average of three determinations for the same dose rate to cells (±SE).

Average values with standard deviation were determined from three independent experiments using different rMnSOD-Lp-CC preparations.

Immunocytochemistry

Tumor cells were incubated for 3 h in the presence or in the absence of rMnSOD-Lp, with the cisplatin alone or with the rMnSOD-LP-CC at concentrations of 92 μm. Subsequently, a cytospin was used to obtain a smear of cells on slides. The smears were fixed in Zamboni solution (4% paraformaldehyde, 15% picric acid) for 60 min and then washed with 1× PBS and incubated for 5 min with 3% hydrogen peroxide to suppress the action of endogenous peroxidase. The immuno-staining was performed using the Dako LSA+ System HRP kit DAKO, CA, USA. The incubation was performed for 30 min with primary antibody anti-rMnSOD-Lp (1:200), produced in rabbit. This step was followed by sequential 30-min incubations with the provided biotinylated link antibody and peroxidase-labeled streptavidin. To complete the reaction, a substrate-chromogen solution was used. The smears were counterstained with hematoxylin.

Immunogold

Both tumor and normal cells were incubated for 6 h in the presence or in the absence of rMnSOD-Lp or with the rMnSOD-LP-CC leader peptide at a concentration of 92 μm. Then normal and tumor cells were fixed with 0.1% glutaraldehyde and 4% paraformaldehyde in 0.1 m sodium cacodylate buffer for 60 min at room temperature and then washed twice in 0.1 m sodium cacodylate buffer. Samples were then treated with 1% OsO4 for 10 min, dehydrated, embedded in Epon 812 and finally the polymerization took place in a heater at 60 °C for 24 h. Ultrathin sections were prepared using the Leica Ultracut UCT (Milan, Italy) ultramicrotome and mounted on nickel grids. The sections were subsequently subjected to antigen unmasking in citrate buffer, incubated with 10% hydrogen peroxide for 10 min, washed three times in PBS 0.9% for 5 min and incubated in BSA 1% and glycine 0.15% in PBS for 30 min. The samples were incubated with polyclonal anti-rMnSOD (rabbit anti-human) diluted (1:20) in Tris–HCl 0.05 m with 1% BSA for 3 days at 4 °C, washed three times in PBS 0.9% for 10 min and incubated with secondary antibody DAR (donkey anti-rabbit) conjugated to 15-nm colloidal gold diluted (1:10) in 0.1% BSA for 2 h at room temperature. The sections were washed in PBS (pH 7.4) and distilled water prior to counterstaining with uranyl acetate and lead citrate. Ultrathin sections were examined using a LEO 912AB Zeiss transmission electron microscope.

Results

Clonogenic assay of normal (MRC-5) and tumor (HTB-112) cells treated in the presence or in the absence of rMnSOD-Lp-CC and CC

The clonogenic tests on normal cells MRC-5 in the presence of the cisplatin alone showed a cell growth inhibition of 80%, at the maximal dose of 0.5 μm.

When the MRC-5 cells were treated in the presence of 0.5 μm rMnSOD-Lp-CC, no inhibition of cell growth was observed (Figure 1).

Figure 1.

 Clonogenic test. On the normal cells MRC-5, the effect of the rMnSOD-Lp-CC at the concentration of 0.5 μm is minimal. In contrast, the same cells treated in the presence of cisplatin alone at same concentration showed a growth inhibition of 80%. The effect of cisplatin alone on HTB-112 cells is minimal. At the concentration of 0.5 μm, cell growth inhibition is only 20%; thus, 80% of tumor cells resist this treatment. In contrast, in the same cells treated in the presence of rMnSOD-Lp-CC at 0.5 μm, the growth inhibition is about 90%.

On the tumor cells HTB-112, the treatment in the presence of 0.5 μm rMnSOD-Lp-CC produced a cell growth inhibition of 97%, while the growth inhibition of HTB-112, in the presence of cisplatin alone at the same concentration, was 40% (Figure 1).

Atomic absorbance spectrophotometric analysis

The presence of Pt in cancer cells following treatment of target cells (about 4.5 × 105) in the presence of cisplatin alone or rMnSOD-Lp-CC was evaluated by atomic absorbance spectrophotometry. HTB-112 cells were incubated with 11.1 μg of Pt (92 μm) in both treatments (CC and rMnSOD-Lp-CC treatment). The results, represented in Table 1, show that the rMnSOD-Lp-CC delivered approximately four times more cisplatin into HTB-112 cells than the amount delivered into cells when cisplatin was used alone.

Table 1.   Quantitative analysis of platinum (Pt) detected by atomic adsorbance analysis on HTTB-112 cell line
SamplesPt (μg/L) detected in pellet of cells% Pt uptake
  1. Data were obtained from three independent determinations performed in the same conditions.

CTRL (cells not treated)00
CP (cells treated in the presence of cisplatin alone)4691.69
LP-CC (cells treated in the presence of rMnSOD-Lp-CC)18566.68

Light microscopy

In endometrial cells (HTB-112), treated with the leader peptide for 3 h, the percentage of labeled cells was observed to be about 20%, most of the cells being in mitosis (M- phase) (Figure 2A,B). After treatment with rMnSOD-LP-CC, for 3 h, a higher percentage of labeled cells (72%) was observed (Figure 2C,D). Cell shrinkage, a typical preapoptotic feature, was detected in these cells, in contrast to cells treated with the leader peptide alone. The treatment of HTB-112 cells with cisplatin alone induced preapoptotic changes in only a small number of cells. A prolonged treatment (6 h) with rMnSOD-LP showed cytoplasmic positivity in 96% of the EC cells, demonstrating internalization of the rMnSOD-Lp (Figure 2E). When the cells were treated with the prolonged treatment (6 h) of rMnSOD-Lp-CC, 98% of them displayed cytoplasmic positivity and nuclear and cytoplasmic apoptotic fragmentations (Figure 2F). As controls (not treated with rMnSOD-Lp or rMnSOD-Lp-CC), the cells were incubated with their specific medium alone (Figure 2G) or with the anti-rMnSOD-Lp antibody alone (Figure 2H). No positivity was detected in either control.

Figure 2.

 (A,B) Cells treated for 3 h with the rMnSOD-LP, followed by labeling with the anti-rMnSOD-Lp antibody: most of the labeled cells (brown-stained) are in M-phase mitosis. (C,D) Cells treated for 3 h with the rMnSOD-Lp-CC, followed by labeling with the anti-Lp antibody: cell shrinkage, an early sign of apoptosis, can be observed. (E) Cells treated for 6 h with the leader peptide alone and labeled with the anti-rMnSOD-Lp antibody: the cytoplasmic positivity (brown) demonstrates internalization of the Lp. (F) A cell treated for 6 h with the rMnSOD-Lp-CC and labeled with the anti-rMnSOD-Lp antibody: nuclear fragmentation and cytoplasmic blebbing are evident. (G, H) Negative control cells, treated, in the absence of rMnSOD-Lp and rMnSOD-Lp-CC, with their specific medium alone (G) or with the anti-rMnSOD-Lp antibody alone (H).

Transmission electron microscopy

Transmission electron microscopy observations confirmed internalization of the leader peptide in HTB-112 cells treated with the rMnSOD-LP, which showed colloidal gold particles (15 nm) mainly dispersed in the cytoplasm and sometimes near the mitochondrial envelope (Figure 3A,B). Cells treated with the peptide conjugated to cisplatin showed the same distribution of positivity in their cytoplasm as cells treated with the peptide alone, thus demonstrating that the rMnSOD-Lp either alone or linked to cisplatin enters the cells. The internalization of cisplatin was also evident in nuclear and cytoplasmic apoptotic fragmentations (Figure 3C,D).

Figure 3.

 (A,B) HTB-112 cells treated for 6 h with the rMnSOD-Lp alone and labeled with the colloidal gold-coupled anti-rMnSOD-Lp antibody: these particles are uniformly distributed in the cytoplasm. (C,D) HTB-112 cells treated for 6 h with rMnSOD-Lp-CC: nuclear and cytoplasmic fragmentations are evident; D is an enlargement of the cytoplasmic apoptotic body.

rMnSOD-Lp-CC also enters normal cells (MRC-5), as demonstrated by the presence of the 15-nm gold particles distributed in their cytoplasm (Figure 4A,B) following the 6-h treatment, and leads to a lesser degree of both nuclear and cytoplasmic damage than in the EC cells. The nucleus shows a slightly indented shape, and the cytoplasm displays a few signs of vacuolization (Figure 4A,B). The MRC-5 untreated control cells show no labeling at all (Figure 4C,D).

Figure 4.

 (A) MRC5 cells treated with rMnSOD-Lp_CC labeled with colloidal gold-coupled anti-rMnSOD-Lp antibody for 6 h: A slight nuclear indentation and a slight cytoplasmic vacuolization are detectable; the degree of damage appears to be less than in the treated endometrial cancer cells of Figure 3C,D. (B) Enlargement of the box of fig A. The gold particles (15 nm in diameter) are uniformly distributed in the cytoplasm demonstrating that rMnSOD-Lp_CC enters the cell. (C) MRC5 untreated control: no labeling is detectable in either the nucleus or cytoplasm. (D) The lack of labeling is evident in this enlargement of the box of Fig C.

Discussion

The potential anticancer activity of the leader peptide conjugated to cisplatin (rMnSOD-LP-CC) on endometrial carcinoma cells (HTB-112) is significant. We measured the amount of Pt inside the tumor cells after treatment with 92 μm of rMnSOD-Lp-CC or 92 μm cisplatin alone; the analysis revealed that rMnSOD-Lp-CC delivers and releases an amount of cisplatin four times higher than that found by using cisplatin alone. This difference in internalization indicates that cisplatin alone is poorly internalized into tumor cells. It is known that the main feature of a drug is to be able to reach the lesion site it is intended to treat. In this regard, cisplatin, as we have seen, is able to penetrate cancer cells only in small quantities. For this reason, clinicians are forced to inject high concentrations of cisplatin, in the hope that most of the drug can reach the cancer cells. Of course, because cisplatin only has an antireplicative activity, it will also be toxic for normal cells in replication phase, hence generating side effects. The advantage of using this carrier-leader peptide, which allows injecting a minimum dose of cisplatin (compared to traditional therapy where high amounts of cisplatin must be employed), is to not only reduce the amount of the drug and its subsequent side effects, but mainly to transform cisplatin’s generic antireplicative activity into a specific and selective antitumor activity. This greatly improves tumor selectivity when it comes to normal tissue and improves the efficacy of the drug. In short, this could result in the survival of patients.

This hypothesis was confirmed by the clonogenic tests, in which we observed that following the treatment of tumor cells in the presence of rMnSOD-Lp-CC, no resistant cell clones were generated. On the other hand, when the HTB-112 cells were treated with cisplatin alone, as demonstrated by the clonogenic test, almost 80% of the tumor cells produced new clones. Thus, these cells resisted the treatment, this probably being due to the fact that only a minimal amount of cisplatin reached and penetrated tumor cells.

However, it must be noted that the leader peptide, as stated above, is able to penetrate all cells and thus also the normal ones. It is then obvious that an amount of cisplatin will also be released in these cells, generating minimal apoptotic damage. The cells then probably easily recover, as we have demonstrated several times (17).

To explain this apparent contradiction, we can say that the free radicals produced by cisplatin are easily neutralized by antioxidant enzymes (superoxide dismutase, catalase and glutathione peroxidase) in normal cells, which contain normal quantities of these latter. Tumor cells, however, contain low amounts of catalase (15) and so cannot repair the oxidative damage, thus resulting in a specific and selective antitumor effect. Indeed, by treating normal MRC-5 cells in the presence of rMnSOD-Lp-CC, the small amount of cisplatin contained therein was not able to kill these cells.

But one more consideration we have to do. From Figure 1 it is evident that cisplatin alone (CC) is more effective at inhibiting growth of the MRC-5 normal cells than when it is conjugated to the leader peptide (LP-CC). In light of this result, experiments are in progress to explain this contradiction. Although we have not yet understood what molecular mechanism is underlying this contradiction, we know that it has already been observed in another normal cell line, the MCF-10 (17). Experiments are in progress in our laboratory aimed to clarify this contradiction. We are directing these studies in two different directions: the first of them is to verify whether the leader peptide conjugated to cisplatin, once it has been internalized in normal cells, is poorly released; in this case, it would be logical to think that its cytotoxic activity appears limited compared to the action of cisplatin used alone.

The other address has been suggested from a lot of data that are accumulating in the literature, that indicate that the use of leaders peptides, which deliver proteins into mitochondrial compartments (as is indeed the MnSOD), are used to increase the specificity and selectivity of anticancer molecules that target the mitochondria. These peptides, together with substances of therapeutic power, transport and release their toxic cargo that has been linked in the mitochondria (24–26), and it is known that mitochondria of normal cells are structurally and functionally different from those of cancer cells because they have a enzyme structure and a different membrane potential (27). We think that in our case, we could create a similar situation, also because cisplatin is just a drug that preferentially intercalates into mitochondrial DNA (28).

Although the use of rMnSOD-Lp-CC has already been successfully tested on a panel of tumor cells derived from human tumors (17), there was no evidence that in EC it would be effective. The above results demonstrated that this new molecule displays a significant cytotoxic activity in EC also. Data are innovative mainly as they show that this new molecule could be used for the treatment of endometrial metastatic cancer in humans, in which cisplatin is considered the preferred therapeutic treatment choice.

Conclusion

The present study has provided strong evidence to the effect that rMnSOD-Lp-CC deserves consideration as a cancer treatment and in particular for EC. Although the molecular mechanism(s) underlying rMnSOD-Lp-CC and tumor specificity remain(s) to be clarified, our observation raises the intriguing possibility that an exogenously supplied product may also have a therapeutic potential against human EC.

Acknowledgments

We thank Franco Iamunno for his assistance in TEM analysis at Zoological Station of Naples and Raffaele Marano for assistance in the Histolological preparations.

Funding

The present study was partially sponsored by the ‘Lega Italiana Per La Lotta Contro I Tumori di Napoli’.

Ancillary