Hybrid Molecules of Benzylguanidine and the Alkylating Group of Melphalan: Synthesis and Effects on Neuroblastoma Cells

Abstract

The therapy of neuroblastoma relies, amongst other things, on administering chemotherapeutics and radioactive compounds, e.g., the (meta-iodobenzyl)guanidine [¹³¹I]mIBG. For special applications (conditioning before stem cell transplantation), busulfan and melphalan (M) proved to be effective. However, both drugs are not used for normal chemotherapy in neuroblastoma because of their side effects. The alkylating drug melphalan contains a (Cl-CH₂-CH₂-)₂N- group in the para-position of the phenyl moiety of the essential amino acid phenylalanine (Phe) and can, therefore, be taken up by virtually all kinds of cells by amino acid transporters. In contrast, mIBG isotopologs are taken up more selectively by neuroblastoma cells via the noradrenaline transporter (NAT). The present study aimed at synthesising and studying hybrid molecules of benzylguanidine (BG) and the alkylating motif of M. Such hybrids should combine the preferential uptake of BGs into neuroblastoma cells with the cytotoxicity of M. Besides the hybrid of BG with the dialkylating group (Cl-CH₂-CH₂-)₂N- bound in the para-position as in M (pMBG), we also synthesised mMBG, which is BG meta-substituted by a (Cl-CH₂-CH₂-)₂N- group. Furthermore, two monoalkylating hybrid molecules were synthesised: the BG para-substituted by a (Cl-CH₂-CH₂-)NH- group (pM*BG) and the BG meta-substituted by a (Cl-CH₂-CH₂-)NH- group (mM*BG). The effects of the four new compounds were studied with human neuroblastoma cell lines (SK-N-SH, Kelly, and LS) with regard to uptake, viability, and proliferation by standard test systems. The dialkylating hybrid molecules pMBG and mMBG were at least as effective as M, whereas the monoalkylating hybrid molecules pM*BG and mM*BG were more effective than M. Considering the preferred uptake via the noradrenaline transporter by neuroblastoma cells, we conclude that they might be well suited for therapy.

Keywords: (meta-iodobenzyl)guanidine (mIBG); melphalan (M); melphalan benzylguanidine hybrids (MBGs); neuroblastoma; noradrenaline transporter (NAT); organic cation transporter (OCT).

Figure 1

Structural formulas of the compounds described in this study. Line 1: benzylguanidine (BG) and the radiotherapeuticals [¹³¹I]mIBG and [¹³¹I]pIBG. Line 2: phenylalanine (Phe; essential amino acid); melphalan (M), and busulfan (B). Line 3: mBBG and pBBG. Line 4: hybrids with dialkylating entities copied from M (mMBG and pMBG). Line 5: hybrids with monoalkylating entities inspired by M (mM*BG and pM*BG). M: 4-[N,N-Bis(2-chloroethyl)amino]phenylalanine; mMBG: {3-[N,N-Bis(2-chloroethyl)amino]-benzyl}guanidine; pMBG: {4-[N,N-Bis(2-chloroethyl)amino]-benzyl}guanidine; mM*BG: {3-[N-(2-Chloroethyl)amino]benzyl}-guanidine; pM*BG: {4-[N-(2-Chloroethyl)amino]benzyl}-guanidine; B: 5-[(Methanesulfonyl)oxy]pentyl methanesulfonate; mBBG: 5-[3-(Guanidinomethyl)phenoxy]-pentyl methanesulfonate; pBBG: 5-[4-(Guanidinomethyl)phenoxy]pentyl methanesulfonate.

Scheme 1

Syntheses of our two MBG hybrid molecules. Reagents and conditions: (a) Boc₂O (3.1 eq.), aq. NaHCO₃ (xs.), room temp., 5 d; 94% (ref. [26]: 100%). (b) 2 (1.2 eq.), NEt₃ (4.2 eq.), CH₂Cl₂, reflux, 20 h for para-4, 6 h for meta-4. (c) Tf₂O (1.1 eq.), pyridine (1.2 eq.), CH₂Cl₂, 0 °C, 50 min; 64%. (d) 4 (4.0 eq.), K₂CO₃ (2.2 eq.), CH₂Cl₂, reflux, 18 h. (e) F₃CCOOH/CH₂Cl₂ (1:1), room temp., 30 min.

Scheme 2

Syntheses of our two M*BG hybrid molecules. Reagents and conditions: (a) N-Bromosuccinimide (1.0 eq.), dibenzoyl peroxide (2 mol-%), CCl₄, reflux, 20 h. (b) NaN₃ (3.0 eq.), DMF, room temp., 5 h. (c) PPh₃ (2.1 eq.), acetone, room temp., 1.5 h; then H₂O (26 eq.), NEt₃ (1.6 eq.), 2 (1.4 eq.), room temp., 2 d. (d) 10 (6.0 eq.), CuI (1.0 eq.), Cs₂CO₃ (2.0 eq.), isopropanol, room temp., 4 d. (e) PPh₃ (1.1 eq.), hexachloroethane (1.1 eq.), CH₂Cl₂, room temp., 20 h. (f) F₃CCOOH/CH₂Cl₂ 1:1, room temp., 30 min.

Figure 2

Uptake of [³H]noradrenaline (c = 0.1 µmol/L) within 15 min at 37 °C into neuroblastoma cells with different expressions of the NAT (SK-N-SH > Kelly >> LS). This uptake was measured in a PBS⁺⁺ (+5.5 mmol/L glucose) control, designated as “100% uptake” and in the presence of a molar excess of one of the following potential competitors (c = 100 µmol/L): mIBG·HCl (■), M (■), mM*BG·TFA (■), pM*BG·TFA (■), mMBG·TFA (■), and pMBG·TFA (■). Mean value ± S.D based on 3 independent sets of experiments, each of which was performed in triplicate.

Figure 3

MTS viability assays with the neuroblastoma cell lines SK-N-SH, Kelly, and LS after incubations with the indicated compounds at 37 °C for 72 h: Effects of M (left: ■), mM*BG·TFA (centre: ■) and pM*BG·TFA (right; ■) administered in concentrations of 0.01, 0.1, 1, 10, 100, and 1000 µmol/L (corresponding to logarithmic horizontal scales). The extreme left of each diagram depicts the corresponding PBS⁺⁺ (+ 5.5 mmol/L glucose) control (whose MTS assay served to define “100% viability”) and a reference treatment with mIBG·HCl (■), administered only in a concentration of 10 µmol/L. Mean ± S.D; 3 independent experiments, each in triplicate.

Figure 4

MTS viability assays with the neuroblastoma cell lines SK-N-SH, Kelly, and LS after incubations with the indicated compounds at 37 °C for 72 h: Effects of mM*BG·TFA (■) and pM*BG·TFA (■) (left), compared to mMBG·TFA (■) and pMBG· TFA(■) (right), administered in concentrations of 0.01–0.05–0.1–0.5–1–5–10 µmol/L (corresponding to near logarithmic horizontal scale). The extreme left of each diagram depicts the corresponding PBS⁺⁺ (+ 5.5 mmol/L glucose) control (whose MTS assay served to define “100% viability”). Mean ± S.D; 3 independent experiments, each in triplicate.

Figure 5

Real-time proliferation monitoring over 96 h in the presence of M (green line), mM*BG·TFA (red line), and pM*BG·TFA (orange line) compared to controls (PBS ⁺⁺ + 5.5 mmol/L glucose, black line). Following the initial tumour cell seeding (at hour 0), the neuroblastoma cells SK-N-SH, Kelly, and LS were treated 24 h later with 10 µmol/L (final concentration) of M and the M*BG hybrids. Treatment with Triton X-100 1%, inducing maximum tumour cell lysis, was used as a negative control. The cell index was normalised after 24 h when treatment had been accomplished. Mean ± S.D.; 3 independent experiments each in triplicate.

Figure 5

Real-time proliferation monitoring over 96 h in the presence of M (green line), mM*BG·TFA (red line), and pM*BG·TFA (orange line) compared to controls (PBS ⁺⁺ + 5.5 mmol/L glucose, black line). Following the initial tumour cell seeding (at hour 0), the neuroblastoma cells SK-N-SH, Kelly, and LS were treated 24 h later with 10 µmol/L (final concentration) of M and the M*BG hybrids. Treatment with Triton X-100 1%, inducing maximum tumour cell lysis, was used as a negative control. The cell index was normalised after 24 h when treatment had been accomplished. Mean ± S.D.; 3 independent experiments each in triplicate.

Figure 6

Real-time proliferation monitoring over 96 h in the presence of M (green line), mMBG·TFA (dark blue line), and pMBG·TFA (light blue line) compared to controls (PBS ⁺⁺ + 5.5 mmol/L glucose, black line). Following the initial tumour cell seeding (at hour 0), the neuroblastoma cells SK-N-SH, Kelly, and LS were treated 24 h later with 10 µmol/L (final concentration) of M and the MBG hybrids. Treatment with Triton X-100 1%, inducing maximum tumour cell lysis, was used as a negative control. The cell index was normalised after 24 h when treatment had been accomplished. Mean ± S.D.; 3 independent experiments each in triplicate.

Figure 6

Real-time proliferation monitoring over 96 h in the presence of M (green line), mMBG·TFA (dark blue line), and pMBG·TFA (light blue line) compared to controls (PBS ⁺⁺ + 5.5 mmol/L glucose, black line). Following the initial tumour cell seeding (at hour 0), the neuroblastoma cells SK-N-SH, Kelly, and LS were treated 24 h later with 10 µmol/L (final concentration) of M and the MBG hybrids. Treatment with Triton X-100 1%, inducing maximum tumour cell lysis, was used as a negative control. The cell index was normalised after 24 h when treatment had been accomplished. Mean ± S.D.; 3 independent experiments each in triplicate.