Super-Long SERS Active Single Silver Nanowires for Molecular Imaging in 2D and 3D Cell Culture Models

Abstract

Establishing a systematic molecular information analysis strategy for cell culture models is of great significance for drug development and tissue engineering technologies. Here, we fabricated single silver nanowires with high surface-enhanced Raman scattering activity to extract SERS spectra in situ from two-dimensional (2D) and three-dimensional (3D) cell culture models. The silver nanowires were super long, flexible and thin enough to penetrate through multiple cells. A single silver nanowire was used in combination with a four-dimensional microcontroller as a cell endoscope for spectrally analyzing the components in cell culture models. Then, we adopted a machine learning algorithm to analyze the obtained spectra. Our results show that the abundance of proteins differs significantly between the 2D and 3D models, and that nucleic acid-rich and protein-rich regions can be distinguished with satisfactory accuracy.

Keywords: Ag nanowire; SERS; cell culture model; machine learning; spatial resolution.

Figure 1

(a) Schematic diagram for the synthesis of single AgNWs by an electrochemical method. (b) Schematic diagram for the mechanism underlying the unidirectional growth of AgNWs. (c,d) SEM images of an AgNW. (e) SEM image of the Ag ball at the carbon nanoelectrode tip. (f) SERS spectrum (black) of 10⁻⁹ M R6G on an AgNW and Raman spectrum (red) of 10⁻³ M R6G. (g) SERS mapping of a single AgNW modified with 4-MBN. Scale bars: (c) 200 nm; (d) 20 μm; (e) 2 μm; (g) 20 μm.

Figure 2

(a) Schematic diagram for inserting single AgNW into a cell culture model. (b,c) Optical images of a 2D model inserted with an AgNW (b) and a 3D model inserted with two AgNWs (c). (d,e) Contour graphs of Raman spectra along AgNW. (d) A 2D model with 130 spectra, and (e) a 3D model with 57 spectra.(f) Average spectra of 2D and 3D models. The number of spectra used for the average spectrum calculation were 270 (2D) and 200 (3D), respectively. Scale bars: (b) 20 μm; (c) 100 μm and (c) inset 30 μm.

Figure 3

(a) Projection of Raman spectra onto a 2D PCA score space for the 2D and 3D models; 95% confidence ellipses enclosing the projected spectra as dots on the PCA score space are also shown. (b) Loading plot of the PC1 and PC2 for the spectra of 2D and 3D models; the black line represents the loading of PC1, and the red line represents the loading of PC2, the grey and red bands indicate the characteristic Raman shifts for the 2D and 3D models, respectively. (c) Results of the classification prediction for the PCA of the 2D and 3D models using the KNN algorithm.

Figure 4

(a) Importance of Raman shift learned from 803 SERS spectra of 2D model using Random Forest algorithm. The red curve represents the fit curve of the importance. Arrows point out four important Raman bands. The red and green arrows point out the two most important bands, respectively. (b) Average spectrum of all the 803 SERS spectra obtained from 2D model. Arrows point out the four corresponding bands in (a). (c) Raman shifts and tentative assignments of the important Raman bands found by Random Forest. (d) Distribution of proteins and nucleic acids in 2D model using K-means++ algorithm. In the upper panel, the red circle indicates cell nucleus. In the bottom panel, the green circle indicates cytoplasm while the area between two green curves indicates the area of ECM. (e) Distribution of proteins and nucleic acids in 3D model using K-means++ algorithm. The optical microscope shows the morphology of outer layer cells (yellow rectangle) in 3D model. The inset (blue rectangle) represents the protein-rich regions and nucleic acid-rich regions in 3D model after the spectral data were analyzed by the machine learning algorithm. Scale bars in (d) and (e) 20 μm.