Microscopic investigation of" topically applied nanoparticles for molecular imaging of fresh tissue surfaces

Abstract

Previous studies have shown that functionalized nanoparticles (NPs) topically applied on fresh tissues are able to rapidly target cell-surface protein biomarkers of cancer. Furthermore, studies have shown that a paired-agent approach, in which an untargeted NP is co-administered with a panel of targeted NPs, controls for the nonspecific behavior of the NPs, enabling quantitative imaging of biomarker expression. However, given the complexities in nonspecific accumulation, diffusion, and chemical binding of targeted NPs in tissues, studies are needed to better understand these processes at the microscopic scale. Here, fresh tissues were stained with a paired-agent approach, frozen, and sectioned to image the depth-dependent accumulation of targeted and untargeted NPs. The ratio of targeted-to-untargeted NP concentrations-a parameter used to distinguish between tumor and benign tissues-was found to diminish with increasing NP diffusion depths due to nonspecific accumulation and poor washout. It was then hypothesized and experimentally demonstrated that larger NPs would exhibit less diffusion below tissue surfaces, enabling higher targeted-to-untargeted NP ratios. In summary, these methods and investigations have enabled the design of NP agents with improved sensitivity and contrast for rapid molecular imaging of fresh tissues.

Keywords: Raman spectroscopy; SERS nanoantennas; SERS nanotags; biomedical optical imaging; molecular imaging; nanomedicine; nanoparticle; optical tags.

Figure 1. Micro-NMI method

(1) One SERS NP flavor was reacted with DyLight 550 and conjugated to IgG1 isotype control antibodies (mAbs), while a different SERS NP flavor was reacted with DyLight 650 and conjugated to anti-EGFR mAbs. (2) Fresh tissues were stained with an equimolar mixture of EGFR-targeted and untargeted (isotype-control) NPs, followed by (3) a brief rinse in PBS (~2 sec) to wash away any excess NP solution from the surface of the tissues. (4) The tissues were then placed in a cryomold containing OCT (freezing media). (5) The cryomold was snap-frozen in 2-methylbutane chilled in liquid nitrogen (<10 sec) and (6) cryosectioned into 10-μm-thick sections. These sections were then placed on a glass slide, fixed in formalin, and coverslipped. (7) A Leica DMi8 widefield fluorescence microscope was used to obtain bright field images of the tissue sections, as well as fluorescence images of the untargeted NPs (conjugated to DyLight 550) and the targeted NPs (conjugated to DyLight 650). Scale bar represents 25 μm.

Figure 2. Image processing

Bright field and fluorescence images of a tissue section from a tumor xenograft stained for 15 min with an equimolar NP mixture are shown on the left. The average of all line profiles oriented perpendicular to the tissue surface is shown for the fluorescence images of targeted NPs (top row) and untargeted NPs (bottom row). The tissue autofluorescence background was measured as the average intensity in the unlabeled tissue, and was subtracted from the NP profiles (“Background subtraction”). The maximum intensity of the NP intensity profiles was assumed to correspond to the tissue surface, which defines the zero depth on the x axes in “Crop to surface.” Finally, the depth-integrated NP concentrations (area under the curve) were evaluated to verify that the micro-NMI results (“Depth-integrated NP concentration”) agree with the signals obtained with the wide-area NMI technique. See text for more details.

Figure 3. Conventional wide-area NMI is used to verify the accuracy of the depth-integrated ratios calculated from micro-NMI

The ratio of the depth-integrated concentrations of the targeted and untargeted NPs are calculated because they provide a reliable estimate of biomarker expression levels that are much less sensitive to nonspecific effects such as uneven illumination, variable staining concentrations, and heterogeneous tissue properties (e.g. diffusion constants). (A) A fresh tissue specimen is stained with an equimolar mixture of targeted and untargeted NPs, rinsed briefly (~2 sec) to remove excess NPs from the surface, and imaged with a customized wide-area NMI device that produces a raster-scanned image of the tissue surface. A ratiometric image of EGFR-targeted vs. untargeted NPs shows elevated ratios for the A431 tumor xenograft (which overexpresses EGFR) and a ratio of unity (no differential binding of the targeted NP) for the benign tissue. (B) Depth-integrated concentrations (area under the curve) of targeted and untargeted NPs (plotted as a function of depth from the tissue surface) from the micro-NMI experiments are shown for tumor xenografts (left) and for benign tissues (right). Concentrations of targeted and untargeted NPs as a function of depth are shown in the red curve (C_tar) and blue curve (C_untar), respectively. Depth-integrated NP concentrations are calculated for the regions shaded in red and blue as ∫ C_tardz and ∫ C_untardz, respectively. (C) Ratios are plotted as a function of staining time for both NMI (black circles) and micro- NMI (grey triangles), showing good agreement (<5% error). Error bars show the standard deviation from N = 5 experiments for each staining time.

Figure 4

The average of approximately 1000 curves showing the concentration of targeted (red) and untargeted (blue) NPs as a function of depth from the tissue surface after the tumor xenografts (A) or benign tissues (B) were stained for 15 min. Shaded regions represent the standard deviation for approximately 1000 curves collected from N = 3 specimens of each tissue type. The concentration ratios of targeted vs. untargeted NPs as a function of depth are shown in (C–D) for tumor xenografts and benign tissues, respectively. Shaded regions represent the propagation of error from the shaded regions in A–B, respectively.

Figure 5. Comparing micro-NMI for 3 different NP sizes

Three different NP sizes were functionalized with either EGFR or isotype control mAbs. Tumor xenografts were stained with an equimolar mixture of targeted and untargeted NPs for 6 min, using either the 120-nm, 200-nm, or 300-nm diameter NPs. After staining, the specimens were rinsed briefly and snap-frozen for examination with the micro-NMI method. ( A) The average of approximately 1000 concentration curves are shown for the targeted NPs (red line) and untargeted NPs (blue line) as a function of depth for each NP size and are normalized to the initial staining concentration. The initial staining concentrations are different for each NP size to account for differences in the total surface area of the NPs (see text for details). Shaded regions represent the standard deviation for approximately 1000 curves collected from N = 5 specimens for each NP size. The grey dashed lines and grey text indicate the depth at which the concentration of the untargeted NP diminishes to 10% of the initial staining (starting) concentration. (B) The depth-integrated ratios (targeted vs. untargeted NPs) for each NP size. Error bars are based on N = 5 experiments for each NP size.