Sticking to the Problem: Engineering Adhesion in Molecular Endoscopic Imaging

Abstract

Cancers of the digestive tract cause nearly one quarter of the cancer deaths worldwide, and nearly half of these are due to cancers of the esophagus and colon. Early detection of cancer significantly increases the rate of survival, and thus it is critical that cancer within these organs is detected early. In this regard, endoscopy is routinely used to screen for transforming/cancerous (i.e. dysplastic to fully cancerous) tissue. Numerous studies have revealed that the biochemistry of the luminal surface of such tissue within the colon and esophagus becomes altered throughout disease progression. Molecular endoscopic imaging (MEI), an emerging technology, seeks to exploit these changes for the early detection of cancer. The general approach for MEI is as follows: the luminal surface of an organ is exposed to molecular ligands, or particulate probes bearing a ligand, cognate to biochemistry unique to pre-cancerous/cancerous tissue. After a wash, the tissue is imaged to determine the presence of the probes. Detection of the probes post-washing suggests pathologic tissue. In the current review we provide a succinct, but extensive, review of ligands and target moieties that could be, or are currently being investigated, as possible cognate chemistries for MEI. This is followed by a review of the biophysics that determines, in large part, the success of a particular MEI design. The work draws an analogy between MEI and the well-advanced field of cell adhesion and provides a road map for engineering MEI to achieve assays that yield highly selective recognition of transforming/cancerous tissue in situ.

Keywords: Cancer; Cell adhesion; Colon; Dysplasia; Endoscopy; Esophagus; Gastrointestinal.

Figure 1

The general protocol for molecular endoscopic imaging (MEI). (a) The GI tract is lined with a mucosal layer (depicted as a grey wavy line). In the esophagus, this layer, if present, is relatively small while in the colon it is quite significant. This mucus layer imposes a barrier that the probes would need to cross to interact with the surface of the epithelium. (b) To remove the mucus layer and expose the epithelial surface and signature molecules, a suspicious site is washed (perhaps with a mucolytic agent) with the water jet of the endoscope. (c) The probes are delivered at the suspicious site via the spray catheter of the endoscope and allowed to incubate with the tissue. (d) Unbound probes are washed away via the water jet of the endoscope prior to detection. (e) The tissue is examined via the endoscope to quantify the presence of the probes.

Figure 2

Schematic of probes interacting with esophageal/colonic epithelium. At sites of transformation (right two panels), the epithelium may express signature molecules that can serve as targets for probes seeking to identify transforming/cancerous tissue. The probes consist of molecules cognate to the signature molecules, i.e. ligands, which can be unconjugated (top two panels) or conjugated to particles (bottom two panels). The goal of MEI is to design probes and a delivery system that allows in situ discrimination between healthy and transforming/cancerous tissue. The general protocol is described in Fig. 1. For the case of probes consisting of ligands conjugated to particles of appreciable size, the particles will experience a force and torque exerted on the particles by the local flow (depicted in lower right-hand image). The relative strengths of the disruptive force/torque and the adhesive force determines if the particulate probes will adhere to the epithelium and the nature of the adhesion.

Figure 3

The selectivity equation. In general, the goal of MEI schemes is to maximize the distinction between transforming/cancerous tissue and healthy tissue, i.e. to achieve maximal selectivity. At the most fundamental level, selectivity will be dictated by the convective/diffusive transport of the probes and the “deposition/adhesive interactions” of the probes with the tissue. The transport and the “interaction” of the probes will be determined by a variety of factors which can be categorized into target tissue parameters [Item I in the figure], probe design parameters [Item II in the figure], and delivery design parameters [Item III in the figure]. The former set of factors are essentially fixed, while the latter two sets of factors can be controlled and manipulated. The exact design chosen, i.e. the parameters stipulated in II and III, strongly influences the operative biophysics and thus has a significant impact on the ultimate selectivity of the assay. The goal of engineering MEI is to achieve an assay design that yields maximal selectivity. Previous work in cell adhesion [e.g. Refs. , , , , , , , , and 73], and those that bridge MEI and cell adhesion, can provide key insights into this rather complex problem.

Figure 4

Engineering MEI. This figure depicts a flow-chart for engineering MEI. As shown, the process starts with a detailed characterization of the biochemistry of transforming/cancerous tissue relative to normal tissue. The goal of this effort is to identify signature molecules or ligands that recognize pathological tissue to a greater extent than normal tissue despite lacking knowledge of the underlying signature molecule (purple rectangles in figure). If the signature molecule is known, ligands cognate to the signature molecule can be developed (dark green rectangle). The various classes of molecules from which such ligands may be derived include the ones listed in the lighter green rectangles to the right. For ligands where the underlying signature molecule is not known, the ligand is typically a peptide or a lectin. With the identification of ligands and signature molecules, the decision of the targeting chemistry can be made. (The decision points are depicted in yellow diamonds in the figure.) Next the mode of detection of the probes needs to be considered. We’ve broadly divided possible detection techniques into three classes (orange rectangles): Raman spectroscopy, fluorescence (including confocal, fiber optic, and near infra-red), and white light microscopy. Utilization of Raman spectroscopy necessitates the probe being made of Raman scattering particles while utilization of white light microscopy necessitates a probe in the micrometer, or larger, size range. If fluorescence is the planned mode of detection, then a range of options are available including unconjugated ligands tagged with a fluorophore and ligand-conjugated nanoparticles/microparticles containing a fluorophore. For both bright field and fluorescent detection schemes, the particles that could be used include biodegradable particles,, polymersomes,, liposomes, and quantum dots (depicted in the blue rectangles). While it is convenient to depict the design process as linear, as we have done here, in reality there is cross-talk between the various stages of the design process. Note that in this schematic, we have restricted our discussion to techniques wherein the delivery and imaging of the probes is done via an endoscope. Other delivery methods e.g. oral and intravenous administration, and detection techniques, e.g. MRI, are possible but are beyond the scope of this review.