Peptide-conjugated nanoparticles for targeted photodynamic therapy

2.1.1 NPs@PS@ATWLPPR

The neuropilin-1 (NRP-1) is a transmembrane glycoprotein and a coreceptor of the vascular endothelial growth factor receptor (VEGFR). It is involved in the axon guidance, angiogenesis, and immune responses [24]. NRP-1 is overexpressed in many types of cancers such as colon carcinoma, prostate, pancreatic carcinoma, lung carcinoma, melanoma, astrocytoma, and neuroblastoma [25].

NRP-1 has been described as a potential target against glioblastoma [26]. A high NRP-1 expression in a glioblastoma sample is correlated with increased malignancy. In contrast, NRP-1 under-expression is associated to lower cancer stem cell migration and proliferation in vitro, and to reduce tumor growth, in vivo. The enhanced NRP-1 expression has been observed in endothelial cells and is correlated to the development of tumor neovascularization [27, 28]. Vascular targeted photodynamic therapy (VTP) was applied using ATWLPPR heptapeptide as a part of specific NRP-1 recognizing sequence. Linked by a spacer arm (6-aminohexanoic acid, Ahx) to chlorin PS, the peptide accumulated in the tumor tissue and potentiated the photodynamic activity. However, the biodistribution studies conducted on mice demonstrated a rapid uptake of the peptide by the liver and the spleen. Consequently, 2 h after intravenous injection (IV), 85% of the total amount of the compound was degraded in the liver [29]. Since the peptide arm, conjugated to the PS, was responsible for its selectivity, the degradation of this peptide fragment was related to the decrease in the PS accumulation in the tumor tissue [30].

In order to reduce the hepato-splenic clearance and peptide degradation, functionalized silica-based NPs, grafted with ATWLPPR-peptide by hydrophilic polymer, were designed for vessel targeting. The NP was also designed as a magnetic resonance imaging (MRI) contrast agent, since it is composed of a silica shell coupled to polyethylene glycol (PEG), doped with gadolinium oxide (Gd₂O₃). The PS was a chlorin derivative, 5-(4-carboxyphenyl)-10, 15, 20 triphenyl-chlorin (TPC), substituted with a succinimidyl ester. In brief, aminopropyltriethoxysilane (APTES) reacted with PEG or propylenediaminetetra-acetic acid (PDTA) containing Gd₂O₃ and was cycle-hydrolysis to obtain the NPs (referred to as NP-TPC) [31], [32], [33] (Figure 2(A)). The size of the NPs was estimated between 3.3 and 3.8 nm [31, 33]. The coupling of the peptide (referred to as NP-TPC-ATWLPPR) raised the NPs’ size to 4.6 ± 3.8 nm. NP-TPC-ATWLPPR was tested for their affinity to NRP-1 using biotinylated VEGF in competitive binding assays. The IC₅₀ values were estimated to be 27 and 56.6 µM for NP-TPC-ATWLPPR and NP-TPC, respectively. The photocytotoxic effect was tested on human breast MDA-MB-231 cells that overexpress a high level of NRP-1. The calculated LD₅₀ (Light dose) for P-TPC-ATWLPPR was estimated at 2.8–5.0 J cm⁻² when cells were treated with 1 µM of NP-TPC-ATWLPPR (power 0.7 W, irradiance 4.54 J cm⁻²). The LD₅₀ value was close to that obtained by a similar treatment with 1 µM NP-TPC [31, 32]. In vivo, the maximal MRI enhancement was found at 2–7 min, after the IV injection of NP-TPC-ATWLPPR in nude mice. The biodistribution, analyzed 75 min after the IV injection, suggested both renal and hepatic clearance for NP-TPC and NP-TPC-ATWLPPR. However, the renal elimination was higher for NP-TPC. The exposure of the tumor U87 grafted cells to NP-TPC-ATWLPPR, revealed that the NPs targeted the peripheral vessels surrounding the tumors in accordance with the high expression of NRP-1 observed in endothelial cells [31, 32] (Figure 2(B)).

Figure 2:

(A) Schematic presentation of NP-TPC-ATWLPPR, (B) maximal MRI signal intensity after injecting 84.2 μmol of Gd for a body weight of 250 g for cerebral biodistribution and brain tumor tissue selectivity of NP-TPC-ATWLPPR, and (C) overview clinical picture of the device applied on the mice [32] with permission from Elsevier and Copyright Clearance Center.

2.1.2 NPs@PS@KDKPPR

Based on the sequence homology of the natural ligand of NRP-1, VEGF-A165, a screening of several peptides was performed. Among the selected sequences, KDKPPR peptide showed a higher affinity to NRP-1 than ATWLPPR; the inhibitor dissociation constant (Ki) was estimated to be 9.0 × 10⁻⁸ mol L⁻¹ for the former as compared to 1.4 × 10⁻⁴ mol L⁻¹ for the latter [34]. The measurement of the affinity to NRP-1 for the K(P1)DKPPR conjugate (i.e. monocarboxylic tetraphenyl porphyrin, P1COOH, linked to the first ε-NH₂ lysine of KDKPPR) was 6 versus 171 μM for the P1-ATWLPPR complexes [35, 36].

The K(Pyro)DKPPR conjugate, using pyropheophorbide-a (Pyro) as PS, was coupled to PEGylated gold nanorods (AuNRs@PEG) through a thiol-maleimide (MI) click reaction to attain a combined hyperthermia and PDT effect (Figure 3) [37]. The AuNRs were functionalized with PEG to prevent any cytotoxicity. The AuNRs@PEG-MI-K(Pyro)DKPPR, illustrated in Figure 3, possessed a length of about 44 nm and a width of about 8 nm. The photophysical properties of the free Pyro were preserved in AuNRs@PEG-MI-K(Pyro)DKPPR, thus showing an unmodified visible absorption profile, a good fluorescence intensity (ϕ _F  = 0.30 in EtOH versus 0.38 for Pyro) and a preserved ¹O₂ production (ϕ ₀ = 0.40 in EtOH versus 0.51 for Pyro).

Figure 3:

Schematization of AuNRs@PEG-MI-K(Pyro)DKPPR.

Adapted from Youssef et al. [24].

The affinity of the AuNRs@PEG-MI-K(Pyro)DKPPR and the K(Pyro)DKPPR free peptide to recombinant NRP-1 was evaluated by competitive binding assay giving IC₅₀ values of 1.5 and 2.0 µM, respectively. The efficacy of AuNRs@PEG-MI-K(Pyro)DKPPR was tested in vitro on human glioblastoma U87 MG cells. No cytotoxicity was obtained with concentrations up to 30 µM (concentration relative to PS), in contrast to the cells treated with the PS alone. A good photodynamic efficiency was found when cells were treated with 30 µM of AuNRs@PEG-MI-K(Pyro)DKPPR for 24 h and then exposed to light irradiation at 652 nm (Fluence 10 J cm⁻², fluence rate 4.54 mW cm⁻²) with a 67% decrease in cell viability. This result supported the effect of both photodynamic and photothermal (PTT) therapies due to the presence of Gold.

The same methodology was also used to graft K(P1)DKPPR conjugate on multifunctional NP platform, namely AGuIX^@ [38], [39], [40]. The designed NPs were first proposed as nontoxic resonance magnetic agents for their imaging properties [41]. The so-called AGuIX@MI-K(P1)DKPPR NPs, illustrated in Figure 4, were tested for vascular-targeted interstitial photodynamic therapy (iPDT), using human umbilical vein endothelial cells (HUVEC) as an in vitro model and human U87 grafted tumors in rodents [38, 39].

Figure 4:

Schematic presentation of AGuIX@MI-K(P1)DKPPR NPs.

Adapted from Thomas et al. [39].

AGuIX@MI-K(P1)DKPPR NPs had a hydrodynamic diameter of approximately 10 nm, making them particularly suitable for rapid renal elimination [42]. The fluorescence (ϕ _F  = 0.7 for P1COOH and ϕ _F  = 0.1 for AGuIX@MI-K(P1)DKPPR in D₂O) and singlet oxygen (ϕ _O = 0.24 for P1COOH and ϕ ₀ = 0.28 for AGuIX@MI-K(P1)DKPPR in D₂O) quantum yields of P1COOH PS were not altered after the peptide addition, demonstrating that the PS could be photoactivated to produce a photocytotoxic effect in vitro and in vivo.

HUVEC cell exposure to AGuIX@MI-K(P1)DKPPR NPs showed no dark cytotoxicity at PS concentrations up to 10 μM. However, the affinity to NRP-1 was altered as the peptide was coupled to the NPs (19 µM for AGuIX@MIK(P1)DKPPR versus 2 µM for the peptide alone). Similar changes were observed when the affinity of AGuIX@MI-K(P1)DKPPR was assessed for human and rat NRP-1. Using the Biacore technology based on the surface plasmon resonance (SPR), the Ki values were estimated at 0.5 μM for KDKPPR and 4.7 μM for AGuIX@MI-K(P1)DKPPR for human NRP-1. For mice NRP-1 cells, the Ki values were estimated at 8.7 μM for KDKPPR and 25.2 μM for AGuIX@MI-K(P1)DKPPR. The use of a scramble peptide (KRPKPD) revealed no binding to recombinant NRP-1 protein, in contrast to the KDKPPR peptide. The specificity of the AGuIX@MI-K(P1)DKPPR NPs was validated using HUVEC overexpressing NRP-1. The incorporation of AGuIX@-MI-K(P1)DKPPR NPs into the cells was increased twice when compared to the NPs carrying the scramble peptide (i.e. AGuIX@MI-K(P1)RPKPD NPs).

In vivo, the distribution of AGuIX@MI-K(P1)DKPPR NPs was assessed in rodents, rats and mice, 96 h post-IV injection of 4 µmol kg⁻¹ (PS equivalent) corresponding to 55 µmol kg⁻¹ (Gd equivalent). Throughout the experimental delay, no signs of clinical toxicity were observed. Ninety six hours post-IV injection, the fluorescence signal was mainly found in the organs of excretion such as the liver, bladder and kidneys. However, the contrast enhancement of Gd quantification was only observed in the kidneys and bladder. The selectivity of the AGuIX@MI-K(P1)DKPPR NPs to the tumor, compared to the healthy brain parenchyma, was validated in nude rats. This was applied on an orthotopic tumor developed from human glioblastoma U87 grafted cells, followed by MRI (T1 sequences). The selectivity of the AGuIX@MI-K(P1)DKPPR NPs towards the tumor vascular endothelium after IV injection was also assessed in nude mouse model either with dorsal chamber or cranial window. Nontargeting NPs (i.e. AGuIX@P1 without peptide and AGuIX@-MI-K(P1)RPKPD with scramble peptide) were visualized in the blood vessels 1 h after injection but disappeared after 4 h due to the fenestration of the tumor vascular system (Figure 5). In contrast, AGuIX@MI-K(P1)DKPPR NPs were more localized on the tumor vessel walls 1 h after injection and were still present after 24 h.

Figure 5:

In vivo selectivity: selectivity of NPs using a dorsal skinfold chamber model, before and 1, 6, and 24 h after IV injection of AGuIX@P1, AGuIX@MI-K(P1)RPKPD and AGuIX@MI-K(P1)DKPPR ([P1] = 6 μmol/kg). Blood vessels are represented in black and P1 fluorescence in red. On the contrary to AguIX@P1 or AGuIX@MI-K(P1)RPKPD NPs, thanks to the targeting peptide, AGuIX@MI-K(P1)DKPPR NPs were fixed to the vessel walls of the tumor tissue.

All pictures have been taken with the same magnification; scale bar represents 100 μm [38].

In addition, the efficacy of the treatment with the AGuIX@MI-K(P1)DKPPR NPs was studied by iPDT performed on nude rats with a cranial anchor (Figure 6). This vascular targeting strategy decreased the tumor growth and extended the rats’ survival rate from seven days, as in the case of AGuIX@P1 NPs, to 13 days (P < 0.0001). This result was associated with decreased tumor metabolism after treatment.

Figure 6:

Kaplan–Meier curves of control rats or rats treated by iPDT using AGuIX@P1 NPs (in red) or AGuIX@MI-K(P1)DKPPR (in green), considering the percentage of tumors not having reached two times their initial volume at the end point. At least seven animals were used for each experimental group. Statistical analysis was performed using the Log rank test and highlighted a statistically significantly difference between AGuIX@MI-K(P1)DKPPR group and control group (P < 0.0001) and between AGuIX@P1 (P = 0.0025).

Adapted from Gries et al. [39].

In the field of brain tumors, the susceptibility weighted imaging (SWI) has recently been examined for glioma imaging [43, 44]. These recent studies showed that SWI can help in the glioma classification thanks to its high sensitivity to the microhemorrhage and the microvascularization itself, which correlates with the tumor grade. After VTP using AGuIX@-MI-K(P1)DKPPR NPs, the onset of microhemorrhages was rapid, occurring within the first few minutes (Figure 7(A)). Specifically, hemorrhages were concentrated at the tumor periphery as early as 1 h after treatment and tended to resolve within 24–48 h post-treatment (Figure 7(B) and (C)). This observation was consistent with the localization of the NRP-1 protein which was mainly expressed in the vessels of the tumor periphery and sometimes at the stromal level [32, 45]. The vascular impact of the treatment was restricted to the tumor borders and it did not affect the vessels of the healthy brain tissues, validating the selectivity of the targeting using the KDKPPR peptide.

Figure 7:

Formation of microhemorrhages after VTP.

The rat with human glioblastoma (U87) was treated with VTP (40 mW, 8 min 40 s, 20.8 J), 4 h after IV injection of AGuIX@MI-K(P1)DKPPR NPs (1.75 μmol kg⁻¹, PS equivalent). Short (A) and medium (B–C) term monitoring of microhemorrhages was performed by MRI using a magnetic susceptibility sequence (SWI), which is extremely sensitive to venous blood, hemorrhages and iron storage. The diameter of the tumor visible in T2 at pretreatment was reported on the SWI images (in red).

Recently, the designing of NPs coupled to different NRP-1 targeting peptides was performed by two other teams. They used CRGDK and tLyp-1 peptides covalently coupled to the NPs.

2.1.3 NPs@PS@CRGDK

Zhao et al. designed a very interesting multifunctional nanosystem composed of PEG-PCL (polyethylene glycol-poly (α-caprolactone)) NPs encapsulating the IR780 PS and the oxygen depot perfluorooctyl bromide (PFOB), and covalently coupled to the tumor homing peptide CRGDK. IR780 was used for its near-infrared light absorption and high fluorescence imaging capability and PFOB was chosen for oxygen storage.

By measuring the ¹O₂ formation in solution via singlet oxygen sensor green (SOSG), they observed that the presence of PFOB enriched the environmental O₂ and decreased the hypoxia. In MDA-MB-231 human breast cancer cell line, the fluorescence of dichloro-dihydro-fluorescein diacetate (DCFH-DA) was 1.8 times greater for CRGDK-targeted NPs than the untargeted ones. In MCF-7 resistant human breast adenocarcinoma cell line, no difference between the two types of NPs was observed. These results were in correlation with the NRP-1 expression level which is high in MDA-MB-231 cells and low in MCF-7 cells. By adding anti NRP-1-monoclonal antibodies, the cellular incorporation of CRGDK targeted NPs strongly decreased. Multicellular tumor spheroids of HT-29 cells overexpressing NRP-1 receptor were prepared and incubated with the CRGDK-targeted NPs. For the NPs holding PFOB, the DCFH-DA fluorescence was detected in the whole sphere at the depth of 65 nm, whereas it was only observed at the edge of the tumor spheroid for the NPs lacking PFOB. However, no fluorescence was observed when the CRGDK-targeted NPs were incubated in multicellular tumor spheroids of MCF-7 with low expression of NRP-1 receptor. In vivo experiments in BALB/c nude mice bearing MDA-MB-231 tumor were conducted, where it was proved that the CRGDK-targeted NPs accumulated in tumor with a tumor-to-total organs ratio of 25%. After IV-injection, the fluorescence signal of IR780 was at the tumor periphery in the case of nontargeted NPs, however, with the CRGDK-targeted NPs, the signal occupied the whole tumor tissue. Indeed, CRGDK-targeted NPs could spread from the vascularized periphery to the avascular tumor area. Using a hypoxia marker (hypoxia-inducible factor (HIF)-1-α), it was proved that the tumor hypoxia decreased due to the presence of PFOB. To evaluate the phototoxicity, PDT treatment was performed at 808 nm (2 W cm⁻², 20 s) 24 h post-injection. As expected, the best efficiency was obtained with the CRGDK-targeted NPs (Figure 8) [45].

Figure 8:

Tumor growth curves in mice treated in vivo on an MDA-MB-231 tumor model with saline solution, NPs/I with or without laser irradiation, NPs/IP with or without laser irradiation, and CNPs/IP with or without laser irradiation. The values are expressed as mean ± SD, (n = 5, *P < 0.01, **P < 0.005, Student’s t-test).

After 24 h, the tumors in laser treated groups were irradiated by an 808 nm laser with a power of 2 W cm⁻² for 20 s. I = IR780, P = Perfluorooctyl bromide (PFOB), NPs = PEG-PCL (polyethylene glycol-poly (α-caprolactone)), C = CRGDK, V ₀ = Initial tumoral volume before treatment [45].

2.1.4 NPs@PS@tLyp1

In 2016, Jiang and coworkers designed a new targeted drug delivery system consisting of a PS (chlorin e6, Ce6, 3.8% loading capacity) chemically incorporated in the shell of D-α-tocopheryl polyethylene glycol 1000 succinate-poly(lactic acid) NPs (TPGS-PLA NPs). These NPs were surface-decorated with CGNKRTR (tLyp-1), a tumor homing and penetrating peptide, to target NRP-1 receptor. The resulted spherical tLyp-1 NPs possessed an average size of around 140 nm. They encapsulated therein a chemo-drug (Doxorubicin, Dox, 9.56% loading capacity). The NPs were then used for chemo-photodynamic combination therapy of Dox-resistant breast cancer. In this combination therapy, the PS played multiple roles. On the one hand, it induced cell apoptosis by PDT. On the other hand, it disrupted the endolysosome membranes to release the encapsulated chemo-drug directly into cytoplasm for an enhanced treatment efficiency in drug-resistant cancers. The in vitro studies on HUVEC and Dox-resistant human breast adenocarcinoma cells (MCF-7/ADR) revealed a cellular uptake enhancement and a photocytotoxicity improvement of tLyp-1 NPs compared to untargeted NPs (660 nm laser irradiation). The in vivo studies on mice bearing MCF-7/ADR tumors demonstrated the targeting efficiency and the penetrative ability of tLyp-1 NPs. This resulted in a considerable accumulation of the PS and Dox into the drug resistant tumors, and thus a more efficient chemo-photodynamic combination treatment. An increase in the antitumor efficiency was detected with tLyp-1 NPs exposed to Laser (170 mW cm⁻²) for 9.8 min at 660 nm (Figure 9) [46].

Figure 9:

Tumor growth curves showing the antitumor efficiency of PBS, Free Dox, NPs with Ce6, and Dox without tLyp-1 in the surface (NPs), NPs with Ce6 and Dox with tLyp-1 in the surface (tLyp-1 NPs) with (Laser+) or without (Laser−) irradiation (170 mW cm⁻² for 9.8 min, 660 nm).

Adapted from Jiang et al. [46].

Table 1 describes the NPs@PS@peptide systems, targeting NRP-1 receptors, regarding the types of NPs, PSs, and the coupling between them, in addition to the NPs size, excitation wavelength (λ _excitation), fluorescence quantum yield (ϕ _F ), singlet oxygen quantum yield (Δ_O.S) and the results obtained in vitro and/or in vivo.

2.2.1 NPs@PS@RGD

Arginyl-glycyl-aspartic acid (RGD) is a well-known peptide targeting α_vβ₃ integrins.

In 2014, Wang et al. reported the synthesis of lipid coated upconverting NPs (UCNs). The coating was constituted of RGD peptide functionalized by PMAO (poly(maleic anhydride-alt-1-octadecene)) grafted DOPE (dioleoyl l-α-phosphatidylethanolamine) (UCN/RGD-PMAO-DOPE). Dextran merocyanine 540 (MC540) PS was adsorbed onto these NPs by hydrophobic interactions (MC540@UCN/RGD-PMAO-DOPE). These upconverting NPs were monodispersed and presented a size of 20 nm. The amphiphilic lipid polymer coating was used to improve the hydrophilic properties of the UCNs’ surface, and protect the drug, by electrostatic repulsion, from aggregation and leakage during the transport. The in vitro experiments were performed on MCF-7 cells overexpressing α_vβ₃ integrins. ROS, mainly ¹O₂, were more importantly produced by MC540@UCN/PMAO-DOPE decorated by RGD peptide (MC540@UCN/RGD-PMAO-DOPE) as compared to NPs lacking RGD (MC540@UCN/PMAO-DOPE). This was attributed to the role of RGD in enhancing the cellular uptake of the NPs through receptor-mediated endocytosis. The production of ROS proved the successful energy transfer from the UCNs to the MC540 PS. In addition, MC540@UCN/RGD-PMAO-DOPE were capable of decreasing the cell viability to 35% after exposure to 980 nm laser for 30 min, thus revealing a more efficient photodynamic effect as compared to MC540@UCN/PMAO-DOPE (65% cell viability) (Figure 10). These results verified that the efficiency of PDT was enhanced by the presence of the RGD peptide [49].

Figure 10:

PDT treatment efficiency on MCF-7 cells. Untreated cells as control group and cells treated with UCN/PMAO-DOPE NPs, MC540 loaded UCN/PMAO-DOPE NPs or MC540 loaded UCN/RGD-PMAO-DOPE NPs after NIR laser irradiation (980 nm, 30 min).

Adapted from Wang et al. [49].

In 2018, Yuan et al. reported a new platform based on mesoporous silica NPs (MSN). Black hole quenchers (BHQs) were doped in the inner walls of the MSN mesopores (MSN-BHQ). The resulted NPs were coupled to photoporphyrin (PpIX) PS through a disulfide bond (SS) to afford MSN-BHQ-SS-PpIX. To prolong their blood circulation time, PEG was coupled to PpIX onto the NPs (MSN-BHQ-SS-PpIX-PEG). A specific targeting agent, RGD peptide, was then conjugated to this nanoplatform (MSN-BHQ-SS-PpIX-PEG-RGD). The MSN-BHQ-SS-PpIX-PEG-RGD NPs had a spherical morphology with an average size of 50 nm. An oxidation–reduction reaction on this platform stimulated the glutathione (GSH)-mediated release of the therapeutic drug by breaking the SS link. GSH is present in large quantities in the tumor cells, which allowed accelerating the release of the drugs. In vitro experiments were performed on cervical cancer cells (HeLa). The fluorescence of MSN-BHQ-SS-PpIX-PEG-RGD was higher than that of free RGD. The cellular incorporation of MSN-BHQ-SS-PpIX-PEG-RGD was measured in HeLa cells with and without GSH. With extra GSH, many red fluorescent spots appeared in the cytoplasm mainly around the nucleus. When tested on two other cell lines (SCC-7 and COS7), MSN-BHQ-SS-PpIX-PEG-RGD showed low cytotoxicity in the dark and good phototoxicity in the presence of light (Figure 11). In conclusion, MSN-BHQ-SS-PpIX-PEG-RGD selectively targeted the tumor environment due to the presence of RGD. This selective toxicity was reinforced by GSH. The redox response of this nanoplatform in the tumor environment made it an important candidate in anticancer PDT and in tumor imaging therapy [50].

Figure 11:

In vitro cell viability of (a) COS7, (b) SCC-7, and (c) HeLa cells incubated with different concentrations of MSN-BHQ-SS-PpIX-PEG-RGD without light and under 30 min. Data is shown as mean ± SD (n = 4).

Adapted from Yuan et al. [38].

In the same year, Hou et al. synthetized nanodumbbell ZnPc-UCN@lipid@RGD. Hydrophobic UCNs were transferred into water to form UCN@lipid. The UCN@lipid and zinc phthalocyanine (ZnPc) were both encapsulated into polymersome (PS) shell to form UCN@lipid@PS nanodumbbell. The exterior of the polymersome possessed many carboxyl functional groups that were coupled to RGD peptide. The UCN’s core converted the NIR rays into visible ones, thus overcoming the problem of the limited light penetration into tissues. The ZnPc-UCN@lipid@PS had a high drug loading efficiency of 18.03%. The transmission electron microscopy (TEM) showed that the ZnPc-UCN@lipid@PS were of spherical form with a size of 150 nm. Whereas, the dynamic light scattering (DLS) gave an average diameter of 195 nm. After coupling the UCNs to ZnPc, the fluorescence intensity of the UCNs decreased, indicating an energy transfer between the UCNs and ZnPc. Following the excitation at 980 nm, ZnPc-UCN@lipid@PS and ZnPc-UCN@lipid@PS–RGD showed the greatest production of ¹O₂. These tests were carried out using 9,10-anthracenediyl-bis(methylene)dimalonic acid (ABDA) probe. The study of the in vitro cytotoxicity in HeLa cells exposed to different concentrations of NPs, showed a greater biocompatibility in the case of UCN@lipid@PS-RGD (97% of cell viability for 500 μg mL⁻¹ of NPs) as compared to UCN@lipid@PS (90%) and UCN@lipid (65%). This result was credited to the presence of RGD that induced the specific incorporation of the NPs in the cells, and hence decreased the unwanted cytotoxicity (Figure 12(A)). After incubation with Hela cells and excitation with a NIR source (980 nm, 1.5 W cm⁻²), ZnPc-UCN@lipid@PS-RGD showed the lowest tumor cell viability as compared to the untreated control cells and those exposed to the other ZnPc loaded-PS NPs (ZnPc-UCN@lipid@PS, UCN@lipid@PS nanodumbbells) (Figure 12(B) and (C)) [51]. In the same conditions, ZnPc loaded UCN@lipid@PS with and withour RGD showed the best and higher production of singlet Oxygen in comparaison with other couples detailed in Figure 12(C).

Figure 12:

(A) Cytotoxicity on Hela cells of UCN@lipid, UCN@lipid@PS, and UCN@lipid@PS-RGD. (B) Viability of Hela cells treated with 1 PBS (Control group), 2 UCN@lipid@PS nanodumbbells plus NIR laser, 3 ZnPc loaded-PS NPs plus NIR laser, 4 (ZnPc + UCN@lipid)@PS NPs, 5 (ZnPc + UCN@lipid)@PS NPs plus NIR laser, 6 (ZnPc + UCN@lipid)@PS-RGD NPs plus NIR laser (980 nm, 1.5 W cm⁻²). The concentration is 500 μg mL⁻¹. (C) Comparison of ¹O₂ production between control groups and experiment groups.

Adapted from Hou et al. [39].

In 2015, Zhao et al. successfully synthesized a novel multi arm polymeric nanosystem for PDT (RGD-8PEG-IR700). The RGD and the IR700 units were coupled to a PEG arm (8 polyethylene glycol). The hydrodynamic diameter of the nanosystem was 6.6 nm. The in vitro results on spheroid tumor model A375 and SKOV3 cells that express α_vβ₃ integrins, showed a stronger fluorescence of the RGD-8PEG-IR700 NPs as compared to 8PEG-IR700. In contrast to IR700 and 8PEG-IR700, RGD-8PEG-IR700 induced a significant phototoxicity on A375 cells, under excitation at 504 nm with an IC₅₀ value of 57.8 nM. No cytotoxic effect was observed in the dark even with a concentration of 1 μM (IR700 equivalent) (Figure 13) [52].

Figure 13:

Determination of the dose-dependent (A) cytotoxicity and (B) phototoxicity of free IR700, 8PEG-IR700, and RGD-8PEG-IR700 in A375 cells. Light irradiation at 660 nm with 3.5 mW cm⁻² for 30 min fluent rate.

Adapted from Zhao et al. [40].

In 2015, Yuan et al. used the seventh generation poly (amidoamine) (PAMAM-G7, P) dendrimer of 8 nm. The PAMAM was coupled with Ce6 PS in addition to PEG or RGD, to obtain the PEG-P-Ce6 NPs and RGD-P-Ce6, respectively. The Φ_Δ was 2.5 times higher for RGD-P-Ce6 than free Ce6 in water. In vitro experiments were performed in A375 cells (nonpigmented melanoma cell line expressing α_vβ₃ integrins (+)) and NIH₃T₃ cells (mouse fibroblast that do not express α_vβ₃ integrins (−)). Cellular uptake in (+) A375 cells was 4.7-fold superior for targeted RGD-P-Ce6 than nontargeted PEG-P-Ce6 NPs, whereas the incorporation into (−) NIH₃T₃ cells was the same for both NPs. The photocytotoxicity was evaluated with various concentrations in both types of cells by illumination at 660 nm (3.5 mW cm⁻², 30 min). No cell killing was witnessed when A375 cells were treated with Ce6 due to its poor incorporation. RGD-P-Ce6 presented an enhanced phototoxicity in A375 cells compared to nontargeted NPs. As expected, in NIH₃T₃ cells, no difference in the phototoxicity between RGD-targeted NPs and PEG-P-Ce6 NPs was detected. The penetration of RGD-targeted NPs and free Ce6 was evaluated in A375 tumor spheroids. Using the same concentrations, i.e. 200, 400, and 800 nM, RGD-P-Ce6 showed a higher cellular uptake than free Ce6 with ratios of 40.8, 58.7, and 79.3, respectively (Figure 14) [53].

Figure 14:

A375 Cell viability after PDT treatment (660 nm, 3.5 mW cm⁻², 30 min).

Adapted from Yuan et al. [41].

In 2017, Kim et al. [42] synthesized a nanoplatform consisting of C60 coupled to PEG and functionalized by Ce6 PS and cyclic RGD peptide (cyclic CKRGDf, denoted by cRGD). The nanosystem had a diameter of 3–4 nm. In vitro experiments were performed with SKOV-3 (high α_vβ₃ expression (+)) and KB (low α_vβ₃ expression (−)) cells [54]. After exposure to light (670 nm, 5.2 mW cm⁻², 10 min), the highest cell death was observed in (+) SKOV-3 with the targeted NPs. As expected, free Ce6 and nontargeted NPs were less phototoxic in both SKOV-3 and KB cells. The phototoxicity of the targeted NPs was also reduced in (−) KB cells. The in vivo experiments using BALB/c nu/nu female mice with SKOV-3 and KB xenografted tumors proved that the targeted NPs had a lower uptake in KB than in SKOV-3 tumors. Due to the small size of the NPs and the EPR effect, both targeted and nontargeted NPs were incorporated into the tumors. However, the highest uptake was achieved by the targeted NPs in SKOV-3 tumor. After realizing PDT treatment on the cells incubated with targeted NPs (670 nm, 5.2 mW cm⁻², 40 min), the reached tumor volumes were 4.5 and 2.1 times smaller than those obtained in mice treated with Ce6 and nontargeted NPs, respectively.

In 2019, Shi et al. described the design of targeted RPTD/HP NPs made of chemo-drug Dox coupled to PEG, via a ROS-cleavable thioketal link, and a targeting cRGD peptide (cyclic RGDfC). This system also encapsulated hematoporphyrin (HP) PS (Figure 15) and was used for oral tongue squamous cell carcinoma treatment [55]. Due to the presence of both hydrophobic Dox and hydrophilic PEG, the molecules self-assembled to form RPTD/HP NPs with a size of about 180 nm.

Figure 15:

Synthesis of RPTD/HP NPs. Adapted from Shi et al. [55].

In vivo studies were performed in HOEC (low α_vβ₃ integrin expression (−)) and CAL-27 (high α_vβ₃ integrin expression (+)) cells. The targeted RPTD/HP NPs were better incorporated in (+) CAL-27 than in (−) HOEC cells. After illumination (633 nm, 10 min, 100 mW cm⁻²), the ROS were formed, which triggered the cleavage of the thioketal bond. Consequently, the Dox was released and entered the nuclei of CAL-27 cells. Synergistic effects of PDT and chemotherapy was observed with both the targeted RPTD/HP and the nontargeted PTD/HP NPs with an IC₅₀ value of 0.89 and 0.68 µM, respectively. Targeted RPTD/HP NPs displayed higher cytotoxicity and phototoxicity than nontargeted PTD/HP NPs, since they delivered more amounts of Dox and HP into the cells. The in vivo studies were implemented with free Dox, free HP, targeted RPTD/HP NPs, and nontargeted PTD/HP NPs in CAL-27 tumor bearing BALB/c nude mice. All the treatments inhibited the tumor growth to a certain extent when compared to the control. However, the best results were obtained with the targeted RPTD/HP NPs (Figure 16). In addition to that, these NPs displayed a strong effect on tumor angiogenesis [55].

Figure 16:

Antitumor effects of Dox, HP + light, nontargeted PTD/HP NPs, nontargeted PTD/HP NPs + light, targeted RPTD/HP NPs, targeted RPTD/HP NPs + light (633 nm, 10 min, 100 mW cm⁻²), with free HP (4.0 μg/mL), PTD/HP, and RPTD/HP nanoparticles (4.0 μg/mL HP and 1.0 μg/mL DOX) and L = Light.

Adapted from Shi et al. [43].

2.2.2 NPs@PS@iRGD

Internalizing-RGD (iRGD, sequence: CRGDKGPDC) is a disulfide-based cyclic RGD peptide that targets integrin α_vβ₃ receptors. The process of tumor-targeting by the iRGD peptide takes place in several steps: First, iRGD is proteolytically cleaved by binding to the surface of cells expressing α _v integrins (α_vβ₃ and α_vβ₅). This cleavage generates the CRGDK fragment, which then binds to NRP-1 and penetrates deeper into the tumor parenchyma. The affinity of iRGD for α _v integrins, compared to conventional RGD, is in the nanomolar range. Besides, the affinity of the CRGDK fragment is stronger for NRP-1 than for α _v integrins. This is due to the C-terminal exposure of a conditional C-end Rule (CendR) motif (R/KXXR/K). The receptor of this motif was proved to be NRP-1. On this basis, the CendR motif is able to bind to NRP-1, thus activating an endocytotic/exocytotic transport pathway that leads to a deeper penetration into the tumor [56].

In 2015, Yan et al. synthesized new NPs, named iRGD-ICG-LPs, by the thin-layer rehydration process [57]. These NPs were liposome-based (LP) in which the indocyanine green (ICG) PS was encapsulated (ICG-LPs) with an efficiency of 93.32 ± 1.25%. ICG-LPs were then grafted with iRGD peptide. The mean dynamic diameter of iRGD-ICG-LPs was 115.91 nm. The in vitro assays were performed using three different cell lines, HUVECs (high expression of α_vβ₃), 4T1 (high expression of α_vβ₃ and NRP-1), and MCF-7 (low expression of α_vβ₃). After the incubation of these cells with ICG-LPs and iRGD-ICG-LPs, iRGD-ICG-LPs showed 1.86-fold and 1.69-fold higher fluorescence intensity in HUVEC and 4T1 cells, respectively, as compared to ICG-LPs (Figure 17(A)). Due to their lower expression of α_vβ₃, MCF-7 cells exhibited an inferior fluorescence as compared to HUVEC and 4T1 cells. The cytotoxicity studies showed that iRGD-ICG-LPs were biocompatible. Under laser illumination, iRGD-ICG-LPs and ICG-LPs exhibited a stronger cytotoxic effect than ICG alone (P < 0.01) at the same dose of ICG. The amount of ¹O₂ produced without light, but in presence of iRGD-ICG-LPs alone, was 1.91 times smaller than with iRGD-ICG-LPs excited by laser at 480 nm. The coupling of iRGD with ICG-LPs induced a very important PDT-PTT effect, thus demonstrating the specific targeting of the tumors. Using the luminescence of ICG, the in vivo results showed a stronger accumulation of the NPs in the tumors than that in the liver, spleen and other organs (Figure 18(A)–(C)). After light illumination of the cells exposed to iRGD-ICG-LPs, the growth of the tumor was suppressed (P < 0.01) (Figure 17(B)). For these NPs, the amount of the generated ROS was 3.82 times greater than that produced in the tumors treated with PBS alone (control).

Figure 17:

In vivo antitumor effect and safety evaluation.

(A) Geometric mean fluorescence intensity of HUVECs and 4T1 cells from flow cytometric analysis, (n = 3), (**) P < 0.01. (B) Tumor growth curves of nontreated mice or mice receiving iRGD-ICG-LPs, ICG-LPs, free ICG, PBS and laser irradiation within 30 days. (**) P < 0.01. An 808 nm laser at a power density of 1.0 W cm⁻² was used to irradiate these samples for 8 min. Adapted from Yan et al. [45].

Figure 18:

In vivo molecular imaging and biodistribution.

The free ICG, ICG-LPs, or iRGD-ICG-LPs were IV administrated to the 4T1 tumor-bearing mice and the tumors and major organs were imaged with the ex/in vivo imaging system. (A) Fluorescence signal was obtained in tumor sites at 1, 12, and 24 h after IV administration of free ICG, ICG-LPs, or iRGD– ICG-LPs with 0.5 mg/kg equivalent ICG. (B) Ex vivo fluorescence images of major organs and tumors were obtained at 24 h post injection of free ICG, ICG-LPs, or iRGD-ICG-LPs with 0.5 mg/kg equivalent ICG. (C) Semiquantitative analysis of fluorescence intensity for the different organs and the tumor showed much higher signal intensity in the tumor of mice received with iRGD–ICG-LPs than those received with free ICG or ICG-LPs. (**) P < 0.01. Laser excitation: 808 nm, 1.0 W cm⁻² for 10 min. Adapted from Yan et al. [45]. With permission from Elsevier and Copyright Clearance Center.

In 2019, Sheng et al. synthesized a novel nanoscale drug [58]. The basis of this NP was the high-density lipoproteins (HDL) in which ICG was encapsulated to give rHDL/ICG. The encapsulation of ICG in NPs presented several advantages such as the stability in the plasma and the lack of precipitation and aggregation. The iRGD peptide was coupled onto this system to afford iRGD-rHDL/ICG. These NPs possessed a hydrodynamic diameter of 90 nm. The in vitro assays in 4T1 cells overexpressing α_vβ₃ integrins, and the in vivo assays with the 4T1 murine breast cancer model, showed much higher fluorescence intensity for iRGD-rHDL/ICG than that of ICG and rHDL/ICG. This was ascribed to the efficient targeting displayed by iRGD. A strong tumor regression was observed after the treatment with iRGD-rHDL/ICG followed by light illumination at 808 nm (1.8 W cm⁻², 5 min), whereas the tumor continued to grow in the absence of any treatment (Figure 19). The iRGD-rHDL/ICG induced necrotic and apoptotic effects on the tumor tissues due to the ROS generation and, consequently, exhibited a greater photocytotoxicity compared to ICG and rHDL/ICG. In comparison with ICG, iRGD-rHDL/ICG accumulated specifically in tumors, exhibited a higher stability in blood, and showed a slower clearance from the body.

Figure 19:

In vivo PDT efficacy in 4T1 tumor-bearing mice.

(A) Tumor growth curves after IV-injection of different formulations at an ICG concentration of 1.5 mg kg⁻¹ (n = 5). (B) Representative image of tumors from the 4T1 tumor-bearing mice sacrificed after being treated with different formulations with 1.5 mg kg⁻¹ of ICG1. (C) Changes in body weight over the PDT treatment period. (D) Hematoxylin and eosin (H&E) images of the tumor tissue section after PDT treatment with 1.5 mg kg⁻¹ of ICG1. Scale bar is 100 μm. Data are presented as mean ± SD, n = 3; the light excitation is at 808 nm, 1.8 W cm⁻² for 5 min. Adapted from Sheng et al. [46]. With permission from Royal Society of Chemistry.

In 2020, Wang et al. developed biodegradable NPs (iMSN/siRNA + miRNA + ICG) [59]. The mesoporous silica NPs (MSN) had a size of 15 nm. ICG was encapsulated in the MSN, with an efficiency of 91%, to afford MSN-ICG. MSN-ICG-iRGD was obtained after the coating of a lipid layer coupled to iRGD peptide on MSN-ICG. Carboxyfluorescein (FAM)-siRNA were adsorbed onto the MSN-ICG-iRGD to improve the targeting and to induce the apoptosis of the cancer cells. After light excitation at 808 nm, ROS were produced, especially ¹O₂, The produced ROS had no significant influence on the genetic silencing activity of RNA, however, the exposure to the irradiation led to undesired heat generation. After irradiation at 808 nm (2.0 W cm⁻² for 5 min), the in vitro results in MDA-MB-231 cells treated with iRGD + MSN/FAM-siRNA + ICG revealed the destruction of the membrane of the endosomal vesicles by the released ROS. iRGD + MSN/FAM-siRNA + ICG induced cancer cell death due to the successful targeting of iRGD and siRNA. In vivo, a tumor of MDA-MB-231 cells expressing galectin-8-YFP (Gal8) was grafted in mice. After light illumination at 808 nm, (2.0 W cm⁻², 5 min), a significant fluorescence of iRGD + MSN/FAM-siRNA + ICG was observed in the tumor and the liver but not in other organs. The results showed that iRGD + MSN/FAM-siRNA + ICG displayed a strong tumor regression (Figure 20).

Figure 20:

Representative images and weight of the isolated tumors from different groups.

**P < 0.01 versus all of the groups, (1) saline, (2) iMSN/NC + ICG, iMSN/Plk1 + 200c + ICG (−light), (3) iMSN/Plk1 + NC + ICG, (4) iMSN/200c + NC + ICG, (5) MSN/Plk1 + 200c + ICG, (6) iMSN/Plk1 + 200c + ICG (+light), and (7) iMSN/Plk1 + 200c + ICG (−light) with. 1 mg/kg siPlk1, 1 mg/kg miR-200c and ICG 720 μg/kg. Light excitation: 808 nm, 2 W cm⁻², for 5 min. NC is a negative control, siRNA nonspecific to any human gene, Plk1 is a Polo-like kinase 1, 200c is a miR-200c mimic. Adapted from Wang et al. With permission from American Chemical Society.

2.2.3 NPs@PS@c-RGD

In 2012, Zhou et al. described the synthesis of UCNPs (NaYF4:Yb/Er) coupled to chitosan, cRGD targeting peptide (cyclic RGDyK, named c(RGDK)), and Pyro PS [60]. The diameter of the resulted UCNP-Pyro-cRGD was 53 nm. U87-MG (α_vβ₃ integrin positive (+)) and MCF-7 (α_vβ₃ integrin negative (−)) cells were used for the in vitro studies. The targeted UCNP-Pyro-cRGD displayed a high affinity for U87-MG cells. An excess of the free cRGD peptide decreased this affinity, demonstrating the receptor-mediated endocytosis. After light illumination (980 nm, 5, min, 500 mW cm⁻²), MCF-7 cells treated with UCNP-Pyro-cRGD were intact, while on the contrary, the cell viability of the treated U87-MG cells was drastically decreased (Figure 21).

Figure 21:

PDT effect on (−) MCF-7 and (+) U87-MG cells treated with different concentration of 0, 50, 100, 200 μg/mL NPs (980 nm, 500 mW cm⁻² for 5 min).

Adapted from Zhou et al. [47].

In 2015, the same team described in vivo studies applied in nude mice bearing U87-MG tumors. Two excitation types were chosen; i.e. 635 nm or continuous-wave at 980 nm, both at 500 mW cm⁻² during 60 min with 1 min interval after each minute of irradiation. Two PDT treatments were performed three days apart. In order to mimic the environment of a deep tumor, they used slices of pork to absorb the light. As expected, the highest phototoxicity was achieved when the tumor was treated with UCNP-Pyro-cRGD and NIR-illumination at 980 nm. This result was comparable to that obtained with Pyro and visible-light illumination at 635 nm [61] (Figure 22).

Figure 22:

In vivo PDT treatment with same concentration of pyro (10 µM). (A) time-dependent tumor growth rate, 14 days after the PDT treatment (B) mice survival rate after different kinds of PDT treatment with light excitation at 635 nm for Vis and 980 nm for NIR (500 mW cm⁻²) for 1 min.

Adapted from Zhou et al. [48].

In 2018, Tang et al. focused on the development of NaScF4: 40% Yb, 2% Er@CaF2 UCNPs. Human serum albumin (HSA) was covalently coupled to the NPs. Ce6 chelating Mn²⁺, for PDT and MRI, was loaded onto the HSA. Finally, the thiolated targeting c(RGDyK) peptide was coupled to the NPs to afford rUCNP@HSA(Ce6-Mn)-cRGD. In vitro studies were performed with human glioma U87 and rat glioma C6 cells. To detect ROS, SOSG and dichloro-dihydro-fluorescein diacetate (DCFH-DA) were used in solution and in cells, respectively. It was proved that upon excitation, a light resonance energy was transferred from rUCNPs to the Ce6-Mn complex leading to 1O2 generation. The highest uptake was observed for the targeted rUCNPs@HSA(Ce6-Mn)-cRGD due to the presence of HSA that could enhance the accumulation through the gp60 receptor and the cRGD peptide that targeted α_vβ₃ integrin. 24 h post treatment, the free Ce6 and the non-argeted rUCNPs@HSA(Ce6-Mn) were eliminated from the tumor, whereas the rUCNPs@HSA(Ce6-Mn)-RGD was still there, showing a great targeting and hence a superior retention ability. The cells were illuminated with a 980 nm laser (1.5 W cm⁻² for 30 min with a 5-min interval between each 5 min illumination) after 6 h of treatment with nontargeted rUCNPs@HSA(Ce6-Mn) or targeted rUCNPs@HSA(Ce6-Mn)-RGD. A strong phototoxic effect was observed with the targeted NPs. In vivo experiments were performed in U87 tumor-bearing mice. After 12 h of incubation with the different compounds, the tumors were illuminated with a 980 nm laser every two days for 14 days (1.5 W cm⁻² for 30 min with a 5-min interval between each 5 min illumination). The group treated with rUCNPs@HSA(Ce6-Mn)-RGD displayed the lowest tumor growth rate. The median survival times for the mice treated by PBS (control), light only, rUCNPs@HSA(Ce6-Mn) or rUCNsP@HSA(Ce6-Mn)-RGD were 45.0, 51.0, 54.5, and 59.2 days, respectively (Figure 23) [62, 63].

Figure 23:

(A) The relative tumor volume after treatment with PBS (control), light only, rUCNPs@HSA(Ce6-Mn) or rUCNsP@HSA(Ce6-Mn)-RGD, with a concentration of 5.2 mg/kg (Ce6 equivalent) and (B) Kaplan–Meier survival time curve. Light excitation at 980 nm (1.5 W cm⁻²) for 5 min.

Adapted from Tang et al. [49, 50].

In 2019, Kohle et al. studied the modification of the diagnostic Cornell prime dots (C′ dots) by encapsulating (design 1) or coupling (design 2) a Methylene blue derivative (MB2) PS. In both designs, the functionalization with a c(RGDyC) targeting peptide was performed. The TEM images presented a diameter of about 4.0 nm for each design. They estimated, respectively, 17 and 14 c(RGDyC) units per MB2 molecule for design 1 and 2. The ϕ ₀ were determined using the singlet oxygen sensor 1,3-diphenylisobenzofuran (DPBF) and were found to be 111 ± 3% for design 1 and 161 ± 5% for design 2. The PS photostability was better in design 1 than in design 2. Surprisingly, the coupling of c(RGDyC) led to a decrease of ϕ ₀ by 25 and 12% for design 1 and design 2, respectively. No in vitro or in vivo studies were performed [64].

In 2005, Kopelman et al. presented a combination of a NP consisting of a polyacrylamide (PAA) core, a cloaking PEG coat, a Photofrin^® PS, a cRGD targeting peptide (cyclic CDCRGDCFC) and an MRI contrast agent, as shown in Figure 24. The size of PAA NPs was about 30–60 nm. The ¹O₂ production was determined using anthracene-9,10-dipropionic acid disodium salt (ADPA) probe. In vitro studies were performed in 9L rat gliosarcoma cells incubated with or without different concentration of NPs. A concentration of 1050 mg mL⁻¹ was required to obtain a PDT effect. Rats bearing intracerebral 9L tumors were used for the in vivo studies. Diffusion-weighted MR images were obtained from untreated rats and those treated either with laser alone or with laser and NPs. Only the latter showed necrosis of the tumor. The final step was the synthesis of targeted NPs using the cRGD peptide and testing them in vitro on (+) MDA-435 and (−) MCF-7 cells. The authors observed that targeted NPs bound only to (+) MDA-435 cells [65].

Figure 24:

The schematic nanoplatform consisting of PAA core matrix with PEG cloaking coat, photodynamic dye (Photofrin^®), MRI contrast enhancement agents and molecular targeting (cRGD peptide).

Adapted from Kopelman et al. [52].

2.2.4 NPs@PS@Fibronectin targeted-peptide

In 2013, Halig et al. formulated iron oxide (IO) NPs encapsulating phthalocyanine 4 (Pc4) PS. These NPs were conjugated to a targeting fibronectin mimetic peptide (Fmp: WQPPRARI), which is well-known to bind to α_vβ₃ integrin overexpressed in head and neck squamous cell carcinoma (HNSCC). Only in vivo multispectral imaging was achieved on mice bearing M4E cell induced tumor. These indicated a high accumulation of the nontargeted IO-Pc4 and the targeted Fmp-IO-Pc4 NPs in tumors as compared to Pc4 alone. No clear effect of Fmp was observed [66].

In 2014, the same team estimated the size of the targeted Fmp-IO-Pc4 NPs of about 41 nm. In vitro experiments were carried out on 4 HNSCC cell lines (M4E-15, TU212, 686LN, M4E CNT). The most significant tumor regression was observed 48 h after laser treatment at 672 nm (100 mW cm⁻², 30 min) for the cells exposed to Fmp-IO-Pc4 NPs as compared to free Pc4, IO-Pc4 NPs, and IO NPs. In vivo experiments were then carried out on HNSCC xenografted mice. The initial tumor size in the mice was 5–7 mm³ before the administration of the targeted and nontargeted NPs and the laser irradiation. 48 h post-administration, the results showed that both NPs prompted a reduced tumor growth compared to free Pc4, but still led to a final increase in tumor volume compared to the initial volume (Figure 25) [67].

Figure 25:

Inhibition of xenograft tumor formation by Pc4 PDT treatment delivered by IO NPs. (A–D) Tumor growth and representative images of tumors on both sides of the mice in (A) PBS control, (B) free Pc4, (C) IO-Pc4, and (D) Fmp-IO-Pc4 groups, respectively with 0.4 mg/kg Pc4. Laser treatment was performed at 48 h post administration (672 nm, 100 mW/cm² for 30 min). Three out of six mice from each group are shown as representatives.

Adapted from Halig et al. [54]. With permission from American Chemical Society.

In 2020, Y. Wang et al. described the synthesis and characterization of Pep-SQ@USPIO nanoprobe used for imaging-guided PDT of triple negative breast cancer (TNBC) [58]. This new cathepsin B (CTSB)-activatable nanoprobe was designed to achieve both fibronectin-targeting magnetic resonance (MR) imaging and near infrared fluorescence (NIRF) imaging. SQ (squaraine-based) PS, known for its high NIR emission and photodynamic effect, was firstly synthesized. The PS was further conjugated with fibronectin-targeting peptide (CREKA) by means of the CTSB-cleavable peptide (GFLG) to form the Pep-SQ conjugate. The Pep-SQ@USPIO nanoprobe was finally obtained by the covalent coupling of the Pep-SQ conjugate onto the maleimide-DSPE-PEG2000-coated ultrasmall superparamagnetic iron oxide (USPIO) NPs. The hydrodynamic diameter of the NPs was 20 nm, as measured by DLS. In vitro PDT testing of the Pep-SQ@USPIO was done on MDA-MB-231 cells overexpressing fibronectin. After 24 h of cell incubation with Pep-SQ@USPIO, the irradiation was applied (5 min, 660 nm, 1 W cm⁻²). A significant decrease in the cell viabilities was induced by the increase of the laser power densities at 660 nm. In addition, cells treated with various concentrations of pep-SQ@USPIO, ranging between 100 and 800 µg mL⁻¹, showed different degrees of apoptosis, consequently verifying the PDT efficiency of this system. The assessment of the accumulation of these NPs by Prussian blue staining revealed a low amount of iron in the heart, lungs, and kidneys, thus demonstrating the absence of pep-SQ@USPIO accumulation. However, the evident Prussian blue staining that occurred in both liver and spleen can arise from both nanoprobe accumulation and the endogenous iron. The obtained results of the NIRF imaging proved that the sufficiently high CTSB activity in the TNBC tumors enabled its detection by Pep-SQ@USPIO. The photodynamic efficiency of Pep-SQ@USPIO was also tested in vivo on MDA-MB-231 tumor-bearing mice. The tumor volume and mass decreased sharply when treated with Pep-SQ@USPIO in the presence of Laser irradiation (660 nm, 2 W cm⁻²) as compared to those exposed to Pep-SQ@USPIO alone, control (PBS), and laser alone (PBS + L). Therefore, it was validated that the Pep-SQ@USPIO activated nanoprobe exhibited a high photodynamic efficiency under laser irradiation, which allows for an enhanced PDT guided by NIRF/MR bimodal imaging for the treatment of TNBC.

In 2021, H. Cao et al. described the elaboration of BKC-NPs (6 nm) formed by the self-assembly of the multifunctional peptide BP-FFVLK-CREKA (BKC) in water. The NPs were constituted of three motifs; hydrophobic BP (bis(pyrene)) PS used for imaging and ROS production, FFVLK peptide for stabilizing the structure, and CREKA peptide for fibrin-targeting on the extracellular matrix of tumors and new-born blood vessels [59]. Using ABDA ((9,10-anthracenediyl-bis(methylene)dimalonic acid)) as ROS detection reagent, the authors observed the formation of singlet oxygen after the two-photon excitation of the BKC-NPs at 800 nm. The in vitro experiments were performed on MCF-7 and HUVECs that express fibrous protein onto their surfaces. No specificity was observed for MCF-7. Whereas, the robust interaction between BKC-NPs and the HUVECs was established by the strong fluorescence that was detected both inside and at the surface of these cells. This interaction was attributed to the abundant expression of fibrous protein onto the surface of HUVECs which allowed for a more specific recognition by the CREKA motif in the NPs. Cells incubated with the same concentration of BKC-NPs were tested under different conditions. After exposure to light illumination at 800 nm, the cell viability attained 20.8%, whereas it reached 16% for one photon (405 nm) and 73% for the control (two-photon laser of 800 nm (1 W) or a xenon lamp with a 405 nm cut-off filter for 20 min). In vivo experiments were then performed using BKC-NPs in BALB/c mice imaging. Confocal laser scanning microscopy revealed blood vessels in the ear and in the tumor. To explore the precision of the therapy in vivo, a selected region was illuminated by a two-photon laser for about 2 min. It was observed that the ROS generated by the accumulated BKC-NPs caused quick vessel breaking as compared to the control vessel that remained intact. It was proven that these new NPs could be used in precise surgery in the brain and eyes for example. Concerning PDT experiments, the mice were injected with these NPs and were illuminated with a two-photon laser at 800 nm at different time intervals. At day 16, it was observed that the tumor volumes increased to attain 695.8, 780.2, and 610.3 mm³ for the control, laser, and BKC-NPs alone, respectively. However, an important tumor volume regression, reaching 67 mm³, was achieved by the BKC-NPs in the presence of irradiation.

2.2.5 NPs@PS@RGD-4R

In 2019, Dai et al. described the elaboration of polymeric NPs consisting of 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[maleimide(polyethylene glycol)-2000] (MPD) matrix encapsulating a fluorogen (TTB). TTB displays an important aggregation-induced emission with strong NIR fluorescence and can efficiently produce ROS especially ¹O₂. These MPD/TTB NPs were covalently coupled, via click reaction, to a targeting peptide, either RGD or RGD-4R (i.e. modular peptide RGDFGGRRRRC), to produce targeted RGD-MPD/TTB and RGD-4R-MPD/TTB NPs. ROS production was demonstrated using dichlorofluorescein (DCFH), dihydrorhodamine 123 (DHR123), and EPR spectroscopy to identify O₂ ^•− and ABPA to detect ¹O₂. The targeted NPs presented a high NIR luminescence at 730 nm and were efficient for real-time fluorescence monitoring. SKOV-3, HeLa, and PC3 cell lines having high α_vβ₃ integrin expression and MCF-7 cell line with low α_vβ₃ integrin expression were used to prove the specificity of RGD-targeted NPs to α_vβ₃ integrins. The incorporation of the targeted NPs in MCF-7 was negligeable, while it was significant in the other cell lines. The targeted RGD-4R peptide presented a higher affinity than RGD. The PDT performed on cells incorporated with targeted RGD-4R-MPD/TTB NPs (730 nm, 200 mW cm⁻², 10 min) revealed an apoptotic rate of 87% for PC3, 89% for HeLa, and 91% for SKOV-3 cells. In comparison, it was only 17% for MCF-7 cells. The xenografted tumor model with HeLa, PC3 and SKOV-3 cells were used for the in vivo studies. A remarkable tumor growth delay was observed after the injection of targeted RGD-4R-MPD/TTB NPs and illumination at 730 nm (200 mW cm⁻², 10 min) (Figure 26). This was accompanied with an inhibition of the expression of BCL2 and Ki-67 genes, and hence led to a reduction in the tumor proliferation and an advancement of apoptosis [68].

Figure 26:

RGD-4R-MPD/TTB (10 μg/mL) NPs mediated PDT for multiple xenograft tumors.

Tumor volume changes in (A) HeLla, (B) PC3, and (C) SKOV-3 cells. Light excitation at 730 nm, 200 mW cm⁻² for 10 min. Adapted from Dai et al. [55].

2.2.6 NPs@PS@RGDfK

In 2014, Haedicke et al. developed calcium phosphonate NPs, in which they incorporated 5,10,15,20-Tetrakis(3-hydroxyphenyl)chlorin (mTHPC) PS. The authors covalently coupled both a targeted RGD peptide (RGDfK) and a fluorescent dye (DY682) for near-infrared fluorescence (NIRF) emission. In vitro experiments were performed using tongue-squamous epithelium carcinoma cells CAL-27. A perinuclear localization of the NPs was observed. In vivo experiments were conducted in CAL-27 xenografted female athymic nude mice. The best accumulation for the nontargeted and the RGD targeted NPs was found at 8 and 24 h post-treatment, respectively. A better internalization in the tumor was observed using RGD targeted NPs in comparison to the non-targeted NPs or Foslip^® PS. However, a strong accumulation was also detected in the lungs, kidneys, spleen, and liver. After PDT (652 nm, 0.1 W cm⁻², 100 J cm⁻²), a decrease of both tumor volume and tumor vascularization was demonstrated in 3 out 4 animals [69].

Table 2 describes the NPs@PS@peptide systems, targeting α_vβ₃ integrin, regarding the types of NPs, PSs and the coupling between them, in addition to the NPs size, excitation wavelength (λ _excitation), fluorescence quantum yield (ϕ _F ), singlet oxygen quantum yield (Δ_O.S) and the results obtained in vitro and/or in vivo.

2.5.1 NPs@PS@HER2 3-340

Narsireddy et al. synthesized 35–40 nm sized multimodal NPs constituting of an Fe₃O₄ core and a chitosan shell. The chitosan-covered Fe₃O₄ NPs were then deposited with gold NPs. 5,10,15,20-tetrakis(4-hydroxyphenyl)-21H,23H-porphine (tHPP) PS was attached via a lipoic acid linker to the gold NPs. A human EGFR2 (or HER2) targeting peptide, known as HER2 3–340 (MGSSHHHHHH SSGLVPRGSH MGVDNKFNKE MRNAYWEIAL LPNLNNQQKR AFIRSLYDDP SQSANLLAEA KKLNDAQAPK), was finally attached to the chitosan shell via a nickel-nitrilotriacetic acid linker. The uptake assays were performed in ovarian SKOV-3 cells. The obtained IC₅₀ values for the PS alone, the nontargeted NPs and the targeted Peptide-NPs were 0.75, 2.1, and 1.7 μM, respectively. The addition of the peptide did not significantly improve the selectivity for HER2. The positive effect of the NPs was demonstrated by the decrease of the dark cytotoxicity compared to the PS alone. The in vivo distribution was studied in FoxN1 nude mice. 24 h after the injection of the nontargeted and the targeted peptide-NPs in the tumors, the estimated gold amount (ng of Au/gm of wet tissue) was found to be 477.5 ± 128 and 1794.2 ± 505, respectively, while a value of 361.4 ± 78 was recorded in the normal tissue. These outcomes validated the advantage of adding a peptide targeting HER2. Twenty four hours after the injection, PDT was performed (PDT-1200 set up, 640–720 nm, 120 J cm⁻², 200 mW cm⁻², 10 min). A notable retarding effect on tumor growth was observed with the targeted NP (Figure 35) [88].

Figure 35:

Tumor growth curve in nude mice. On day 0, PS formulations were injected via tail vein; and on day 1, PDT was conducted. (A) PBS, (B) PS (2 µM), (C1) ADPN (2 µM of PS), and (C2) ADPN (2 µM of PS). Laser irradiation with 640–720 nm (200 mW cm⁻² for 10 min).

Adapted from Narsireddy et al. [74]. With permission from Elsevier.

One year later, the same team synthesized a fourth generation poly(amidoamine) (PAMAM) dendrimer (i.e. DPN and ADPN) coupled to the same PS and the same peptide to target HER2. Using confocal microscopy, the authors observed a greater fluorescence with the targeted ADPN dendrimer than with the nontargeted DPN dendrimer in SKOV-3 cells. PDT was performed on SKOV-3 (+) and human breast cancer MCF-7 cells (−) (PDT-1200 set up, 20 J cm⁻², 50 mW/cm², 6.5 min). The IC₅₀ values were 0.175, 0.100 and 0.075 μM for the PS, the nontargeted DPN dendrimer and the targeted ADPN dendrimer, respectively. In MCF-7 cells, no difference in the IC₅₀ values was detected between the PS and both dendrimers. The in vivo phototoxicity experiments (PDT-1200 set up, 120 J cm⁻², 200 mW/cm², 10 min) revealed a delay in the tumor growth when exposed to the nontargeted DPN dendrimer or to the targeted ADPN dendrimer. However, no photodynamic effect was observed in tumors exposed to PBS (untreated) or PS alone. It seemed that grafting the targeted peptide did not enhance the efficacy of PDT [89].

2.5.2 NPs@PS@GE11

GE11 is a dodecapeptide with 12 amino acids (YHWYGYTPQNVI). Its affinity towards EGFR was identified by Li et al. GE11 has a lower affinity for EGFR (k _d  = 22 nM) than EGF (k _d  = 2 nM). It increases the endocytosis of the NPs due to an alternative EGFR-dependent actin-driven pathway. It has been shown that the EGFR level stays constant after the incorporation of GE11, indicating an EGFR recycling process with a prolonged cell receptivity for the circulating GE11 [90]. In 2012, Master et al. developed PEG-co-polycaprolactone (PEG-PCL) NPs encapsulating phthalocyanine 4 (Pc4) PS and covalently coupled to GE11 peptide ligand to target EGFR overexpressed on many cancer cell lines. For their in vitro studies, the authors selected A431 human epidermoid carcinoma cells and MCF-7 human breast cancer cells overexpressing EGFR. After 1 and 5 h of incubation, a better accumulation into A431 (+) cells was observed for the targeted NPs as compared to the untargeted ones. However, no accumulation was witnessed into the MCF-7 (−) cells. This was in favor of the receptor-mediated internalization of the targeted NPs. Twenty four hours post incubation, no difference was detected between both NPs. This was due to the processes of passive uptake relying on the EPR effect and the active uptake via the EGF receptors. After 200 s of light illumination (diode array, 675 nm, 200 mJ cm⁻²), it was revealed that the cell viability evolves linearly with the amount of the accumulated NPs. The higher the accumulation, the greater the cell death. In the same year [91], the authors optimized various parameters, i.e. the density of the GE11 peptide ligand on the micelle surface, the Pc4 loading and the light dose, in order to enhance the PDT efficiency. They discovered that the micelle formulation should be of 10% mole GE11-modified polymer and 50 μg Pc4 per mg of polymer. In addition, the illumination using the light-emitting diode array (675 nm, 200 mJ cm⁻²) was prolonged to 400 s [92].

Table 5 describes the NPs@PS@peptide systems, targeting α_vβ₃ integrin, regarding the types of NPs, PSs, and the coupling between them, in addition to the NPs size, excitation wavelength (λ _excitation), fluorescence quantum yield (ϕ _F ), singlet oxygen quantum yield (Δ_O.S), and the results obtained in vitro and/or in vivo.

2.6.1 NPs@PS@GX1

In 2017, Guo et al. developed new angiogenic vessel-targeting (AVT) NPs as delivery vehicles for a hypoxia-activated bioreductive prodrug (Tirapazamine, TPZ). The AVT NPs were formed by the self assembly of a TPC-GX1 conjugate. This conjugate was composed of the TPC PS (5-(4-carboxyphenyl)-10, 15,20-tris(3-hydroxyphenyl)chlorin) covalently linked to an angiogenic vessel-targeting cyclopeptide GX1. The elaborated NPs had an average hydrodynamic size of about 109 nm. TPZ was then loaded on the AVT-NPs, at a capacity of 11.08%. The synthesized AVT-NP/TPZ system was designed for application in the chemo-photo synergistic cancer therapy. TPC-PEG micelles and chemotherapeutic cisplatin were used to track the distribution of the AVT-NPs and to demonstrate their targeting capability. The in vitro studies were performed on MCF-7, 3T3 and HUVEC cells. As for the in vivo assays, they were conducted on MCF-7 and MDA-MB-231 xenograft mouse models. After exposure to the He–Ne laser irradiation (650 nm, 1.2 W cm⁻², 10 min), TPC, in the AVT-NP/TPZ system, interacted with the molecular oxygen to produce ¹O₂ that initiated firstly the cancer cell killing by PDT. The PDT-induced hypoxia triggered a promoted angiogenesis, thus resulting in an enhanced targeted delivery of the NPs specifically to the tumor site, in addition to the activation of the bioreductive prodrug TPZ, that in its turn generate highly cytotoxic free radicals for a combinational chemophototherapy (Figure 36). This work showed the great potential of using hypoxia-induced enhancement of angiogenesis to mediate a specific accumulation of the NPs at the tumor site and initiate a chemo-photo combinational treatment [97].

Figure 36:

MCF-7 tumor (A) and MDA-MB-231 (B) growth curves after intravenous injection of different drug formulations (0.1 mg/kg; Ps/body weight).

Error bars indicate standard deviations (n = 5).*P < 0.05, **P < 0.01, ***P < 0.001 (two-tailed Student’s-test). He–Ne laser irradiation (650 nm, 1.2 W/cm² for 10 min). Adapted from Guo et al. [82].

Table 6 describes the NPs@PS@peptide systems, targeting α_vβ₃ integrin, regarding the types of NPs, PSs, and the coupling between them, in addition to the NPs size, excitation wavelength (λ _excitation), fluorescence quantum yield (ϕ _F ), singlet oxygen quantum yield (Δ_O.S), and the results obtained in vitro and/or in vivo.

2.6.2 Peptide targeting p32

The p32 protein is a trimer with three homologous subunits [98]. It is a mitochondrial matrix protein in normal tissues, but it can also be detected on the cell surface, nucleus, and endoplasmic reticulum. p32 is overexpressed on the cell membrane of certain human cancers making it a useful target in tumor diagnosis and therapy.

α-Helix p32 is a stick-shaped peptide expressed in the main cancer types. The coupling of α-helix p32 with anticancer drugs plays a rigorous role in targeting more than 50% of human cancers [99, 100].

2.6.3 NPs@PS@α-helix p32 membrane

In 2015, Zhang et al. described a new approach, called conformational epitope imprinting, to develop new targeted nanocarriers using the three-dimensional conformation of an epitope, rather than its linear structure, for the specific recognition of p32 membrane protein. To achieve this goal, the authors produced the HAPPE peptide (i.e. hybrid apamin-p32 polypeptide). It consisted of a disulfide-linked α-helix-containing peptide, i.e. apamin, which mimics the extracellular structured N-terminal region of the p32 membrane protein, where seven residues were replaced by topologically equivalent ones from p32. For epitope imprinting, they designed molecularly imprinted polymeric NPs (MIPNPs), possessing selective cavities for the specific recognition of p32 membrane protein, using the HAPPE peptide as template. MIPNPs had a particle size of 37 nm. Nonimprinted polymeric NPs (NIPNPs), using a linear analo of HAPPE peptide (i.e. four cysteines of apamin were replaced by alanines), were also synthesized to demonstrate the specific recognition. Compared to NIPNPs, a strong binding interaction of the MIPNPs with p32 was observed. A fluorescence probe, 6-aminofluorescein (FAM), was encapsulated into MIPNPs and NIPNPs. The uptake of both NPs by p32-positive cancer cells, namely 4T1 murine breast cancer and BxPC-3 human pancreatic cancer cells was assessed by flow cytometric measurement. MIPNPs displayed a greater cellular uptake as compared to NIPNPs. The in vivo biodistribution of MIPNPs and NIPNPs, both encapsulating a near-infrared fluorophore (IR-783 dye), was also studied using 4T1-tumor-bearing mice. A higher accumulation in tumors was presented by MIPNPs. The pre injection of Lyp-1 peptide, known to bind specifically to the N-terminal region of p32, significantly reduced this accumulation. This result confirmed that MIPNPs specifically bound to the targeted p32 protein. The in vivo antitumor effect of methylene blue (MB)-loaded NPs (MIPNPs and NIPNPs) was evaluated using laser light irradiation at 650 nm (800 mW cm⁻², 10 min). It was demonstrated that the MB-loaded MIPNPs exhibited an efficient targeted photodynamic therapy (Figure 37) [101].

Figure 37:

(A) In vivo antitumor effect of PDT on 4T1-tumor bearing mice performed using distinct NP formulations. All injections were performed once at t = 0, MIPNPs (1 mg/mL) loaded with the photosensitizer methylene blue (MB, 3 µm). Tumor volumes were measured every other day for one week. (B) Images of mice treated with (a) NIPNPs and (b) MIPNPs.

Scale bar = 5 mm 650 nm laser light irradiation, 800 mW cm⁻², 10 min. Adapted from Zhang et al. [86]. With permission from John Wiley and Sons.

Table 7 describes the NPs@PS@peptide systems, targeting α_vβ₃ integrin, regarding the types of NPs, PSs, and the coupling between them, in addition to the NPs size, excitation wavelength (λ _excitation), fluorescence quantum yield (ϕ _F ), singlet oxygen quantum yield (Δ_O.S), and the results obtained in vitro and/or in vivo.