The transformation of the nuclear nanoarchitecture in human field carcinogenesis

Abstract

Morphological alterations of the nuclear texture are a hallmark of carcinogenesis. At later stages of disease, these changes are well characterized and detectable by light microscopy. Evidence suggests that similar albeit nanoscopic alterations develop at the predysplastic stages of carcinogenesis. Using the novel optical technique partial wave spectroscopic microscopy, we identified profound changes in the nanoscale chromatin topology in microscopically normal tissue as a common event in the field carcinogenesis of many cancers. In particular, higher-order chromatin structure at supranucleosomal length scales (20-200 nm) becomes exceedingly heterogeneous, a measure we quantify using the disorder strength (L_d ) of the spatial arrangement of chromatin density. Here, we review partial wave spectroscopic nanocytology clinical studies and the technology's promise as an early cancer screening technology.

Keywords: carcinogenesis; chromatin; field effect; heterogeneity; nanocytology.

Figure 1.. Partial wave spectroscopic microscope is an epi-illumination bright-field spectroscopic microscope with small illumination numerical aperture, moderate collection and 40× magnification.

Wavelength-resolved image acquisition was performed by using a Xenon white-light lamp illumination followed by spectral filtration of the incident light via an acousto-optical tunable filter. The resulting microscope images (x,y) are obtained at 200 1 nm spaced wavelengths λ of the incident light spanning the spectral range of 500–700 nm, combined into a 3D (x,y,λ) data cube, and used for subsequent spectral analysis. AOTF: Acousto-optical tunable filter; CMOS: Complementary metal-oxide semiconductor; NA: Numerical aperture.

Figure 2.

Pre-neoplastic cells have significant nanoscale physical alterations in the nucleus. Transmission electron microscopy micrographs of rectal cell nuclei of (A) a healthy patient and a (B) patient harboring tumor elsewhere in colon, representing the field effect of carcinogenesis. Scale bar corresponds to 1 μm.

Figure 3.. Representative transmission bright-field microscope images (bottom row) of histologically normal buccal cells from a healthy patient (left) and a patient with lung cancer (right).

Nuclei were selected using the transmission images after which their disorder strength distribution was obtained (top row). Scale bar corresponds to 2 μm.

Figure 4.. Nuclei segmentation of an isolated, representative buccal cell.

(A) Isolated buccal cell imaged with light transmission, (B) maximum entropy thresholding, (C) watershed segmentation and (D) with resulting segmented nuclei outlined.

Figure 5.. Nuclear L_d increases in human field carcinogenesis.

(A) The nuclear L_d was calculated from buccal cells obtained from the oral mucosa of smokers (control) compared with patients harboring lung cancer. Nuclear L_d is increased significantly in buccal cells from cancer patients (effect size: 1.04, p-value < 0.001, n = 38 patients total). Panel (B) shows the nuclear L_d calculated form FFPE prostrate tissue samples from progressors and nonprogressors. FFPE slides were stained with low concentration of H&E to provide contrast. Nuclear L_d is significantly increased in the patients who went on to develop the disease (progressors), compared with the benign form (nonprogressors). Therefore, nuclear L_d can be a useful marker in lung cancer screening, while the prostate nuclear L_d data represent the first test to differentiate between those who will actually go onto present malignant forms of the disease, which will correspond to alterations in biological pathways (i.e., chromatin structure and gene expression). Error bars represent standard error. FFPE: Formalin-fixed and paraffin embedded; H&E: Hematoxylin and eosin.