**Phage Display-Based Homing Peptide-Daunomycin Conjugates for Selective Drug Targeting to PANC-1 Pancreatic Cancer**

**Levente E. Dókus 1,2, Eszter Lajkó 3 , Ivan Randelovi´c ¯ 4 , Diána Mez˝o <sup>3</sup> , Gitta Schlosser 1,2 , László K˝ohidai <sup>3</sup> , József Tóvári <sup>4</sup> and Gábor Mez˝o 1,2,\***


Received: 20 May 2020; Accepted: 17 June 2020; Published: 22 June 2020

**Abstract:** The Pancreatic Ductal Adenocarcinoma (PDAC) is one of the most aggressive and dangerous cancerous diseases, leading to a high rate of mortality. Therefore, the development of new, more efficient treatment approaches is necessary to cure this illness. Peptide-based drug targeting provides a new tool for this purpose. Previously, a hexapeptide Cys-Lys-Ala-Ala-Lys-Asn (CKAAKN) was applied efficiently as the homing device for drug-loaded nanostructures in PDAC cells. In this research, Cys was replaced by Ser in the sequence and this new SKAAKN targeting moiety was used in conjugates containing daunomycin (Dau). Five different structures were developed and tested. The results indicated that linear versions with one Dau were not effective on PANC-1 cells in vitro; however, branched conjugates with two Dau molecules showed significant antitumor activity. Differences in the antitumor effect of the conjugates could be explained with the different cellular uptake and lysosomal degradation. The most efficient conjugate was *Dau*=*Aoa*-GFLG-K(*Dau*=*Aoa*)SKAAKN-*OH* (conjugate **4**) that also showed significant tumor growth inhibition on *s.c.* implanted PANC-1 tumor-bearing mice with negligible side effects. Our novel results suggest that peptide-based drug delivery systems could be a promising tool for the treatment of pancreatic cancers.

**Keywords:** pancreatic cancer; targeted tumor therapy; homing peptide; antitumor peptide conjugates; daunomycin; oxime linkage

#### **1. Introduction**

Pancreatic Ductal Adenocarcinoma (PDAC) is one of the most aggressive and dangerous cancerous diseases with a high mortality rate [1]. In the USA, more than 55,000 new cases were estimated in 2018 which is 3–4% of the all newly diagnosed cancer cases. Approximately 80% of these incidences will lead to death within a year [2]. The average 5-year survival rate is less than 5% [3]. The main reason for the high mortality of pancreatic cancer patients could be very poor prognosis. The early diagnosis of PDAC is still difficult, and most patients have already progressed to not operable and incurable statuses at the recognition of the disease [4]. In addition, the chemotherapy applied to treat pancreatic cancers is usually ineffective due to the fast development of resistance. Chemotherapy causes many side effects

because of the low selectivity of the currently used drugs [5]. Furthermore, the hypovascularity of PDAC and a dense desmoplastic stroma that create barriers restrict drug delivery to the tumor site [6]. Therefore, the design of efficient anticancer agents against PADC is one of the most challenging tasks for scientists working on cancer research [7]. Targeted tumor therapy could be a promising strategy to overcome these drawbacks in pancreatic cancer treatment—similar to other types of cancers [8]. Targeted tumor therapy is based on targeting tumor-specific or overexpressed receptors or other cell surface compartments on tumor cells that can be recognized selectively by antibodies or small molecules like folic acid or peptides [9,10]. Drug molecules attached to these homing moieties can enter specifically into tumor cells, resulting in selective toxicity without causing toxic side effects in healthy tissues. The application of small molecule drug conjugates (SMDCs) over antibody-drug conjugate (ADCs) may have an advantage in the treatment of PDAC because SMDCs have higher tissue permeability [11].

Several homing peptides that recognize pancreatic cancer cells and could be used for drug targeting directly or as a part of nanoparticles have been described in the literature [12–15]. One of them is the CKAAKNK oligopeptide that was selected by phage display technique and which can specifically bind to tumor vessels in RIP-Tag2 transgenic mice, a prototypical mouse model of multistage pancreatic islet cell carcinoma [16]. Valetti et al. attached the CKAAKN homing peptide to a squalene (SQ) molecule via thiol-maleimide Michael addition coupling [17]. The conjugate was co-nanoprecipitated with the squalenoyl prodrug of gemcitabine (SQdFdC) resulting in nanoparticles. The construct was tested on MIA PaCa-2 human pancreatic adenocarcinoma cells, which overexpress frizzled-5 (FZD-5) receptors compared to NIH/3T3 fibroblasts. It was indicated that these cells selectively took up the nanoparticles decorated with the homing peptide by a receptor-mediated way [18]. Frizzled receptors as Wnt binding 7TM GPCRs are key players in the Wnt/β-catenin signal pathway that is commonly hyperactivated in pancreatic cancers, leading to enhanced cell proliferation [19]. In the presence of appropriate ligands, the FZD-5 receptor can be internalized, usually in a heterodimeric form [20]. Therefore, this protein is a promising target for drug targeting to tumor cells. In addition, these data indicated the efficacy of the CKAAKN peptide as a homing device—related to the Wnt-2 sequence—for targeted tumor therapy.

In our research, several SMDCs were developed, derived from the CKAAKN oligopeptide. The structure–activity relationship was investigated as well. In this work, cysteine was replaced by serine to remove the unnecessary thiol group at the conjugation site. This exchange is widely used to eliminate reactive thiol group when it is not essential for the biological activity. In addition, the substitution of Cys by Ser improves the hydrophilicity and solubility of the peptide and its conjugates. Daunomycin (Dau), as an anticancer agent was attached to the homing peptides via oxime linkage, which shows proper stability in the circulation and allows the release of an active metabolite in lysosomes [21,22]. This active metabolite contains an amino acid (Aaa) to which daunomycin is connected through an aminooxyacetyl moiety (*Dau*=*Aoa*-Aaa-OH). This metabolite was proved to bind to DNA; however, the binding efficacy highly depends on the type of the amino acid [22].

The PANC-1 cell line, originally derived from head pancreatic carcinoma, was applied in our studies, and has an invasive phenotype and the ability to give metastasis to the peripancreatic lymph node; thus, this cell line can be considered as an in vitro model of lymph-node-positive PDAC. The invasiveness and metastatic potential of pancreatic cancer cells has been shown to be influenced by the Wnt/β-catenin pathway. This Wnt/β-catenin pathway has also been reported as a central element of immune-escape mechanisms of pancreatic tumors by providing an environment with immune-tolerogenic cytokine and chemokine [23]. Moreover, the expression level of β-catenin, a key protein of the Wnt pathway, has been found to be well-correlated with the gemcitabine-resistance of different pancreatic cell lines, including PANC-1 cells [24]. These three characteristics of pancreatic tumor cells like PANC-1 seem to be interrelated and orchestrated by the Wnt/β-catenin pathway.

The influence of the number of drugs as well as the presence of an enzyme cleavable spacer on antitumor activity was studied in PANC-1 pancreatic cancer cells. The antitumor effect is influenced by several cellular factors; therefore, the binding, the cellular uptake and the metabolism of the conjugates

were also investigated. The best compounds identified in the in vitro studies were applied in vivo experiment using subcutan (*s.c.*) developed PANC-1 tumor-bearing SCID mice. The results were compared with free drug administration.

#### **2. Materials and Methods**
