Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2010 Jun 1;3(3):181-94.
doi: 10.1593/tlo.09310.

Strategies for High-Resolution Imaging of Epithelial Ovarian Cancer by Laparoscopic Nonlinear Microscopy

Affiliations
Free PMC article

Strategies for High-Resolution Imaging of Epithelial Ovarian Cancer by Laparoscopic Nonlinear Microscopy

Rebecca M Williams et al. Transl Oncol. .
Free PMC article

Abstract

Ovarian cancer remains the most frequently lethal of the gynecologic cancers owing to the late detection of this disease. Here, by using human specimens and three mouse models of ovarian cancer, we tested the feasibility of nonlinear imaging approaches, the multiphoton microscopy (MPM) and second harmonic generation (SHG) to serve as complementary tools for ovarian cancer diagnosis. We demonstrate that MPM/SHG of intrinsic tissue emissions allows visualization of unfixed, unsectioned, and unstained tissues at a resolution comparable to that of routinely processed histologic sections. In addition to permitting discrimination between normal and neoplastic tissues according to pathological criteria, the method facilitates morphometric assessment of specimens and detection of very early cellular changes in the ovarian surface epithelium. A red shift in cellular intrinsic fluorescence and collagen structural alterations have been identified as additional cancer-associated changes that are indiscernible by conventional pathologic techniques. Importantly, the feasibility of in vivo laparoscopic MPM/SHG is demonstrated by using a "stick" objective lens. Intravital detection of neoplastic lesions has been further facilitated by low-magnification identification of an indicator for cathepsin activity followed by MPM laparoscopic imaging. Taken together, these results demonstrate that MPM may be translatable to clinical settings as an endoscopic approach suitable for high-resolution optical biopsies as well as a pathology tool for rapid initial assessment of ovarian cancer samples.

Figures

Figure 1
Figure 1
Evaluation of MPM (yellow) and SHG (blue and grayscale) potential for examination of mouse EOC. Normal mouse ovary (A–C), ovarian carcinomas of TgMISIIR-TAg (D–F), and disseminated peritoneal EOC (G–M) mouse models were visualized by H&E staining (A, D, and G). MPM intrinsic emission (B, E, and H) and SHG imaging (B, C, E, F, H, I–M). (A–C) The ovarian surface epithelium (OSE, arrow), part of the corpus luteum (CL), and the ovarian bursa (OB) are all clearly resolvable by MPM and resemble those in conventional histologic image. In addition, SHG demonstrates collagen in the ovarian bursa and the basement membrane (BM) underneath of OSE. (D–F) Monomorphous polygonal neoplastic cells (arrow) are detectable by H&E and MPM. Note that SHG demonstrates presence of collagen with most collagen fibers near tumor periphery being located perpendicular toward its surface (arrowhead), a feature not evident in conventionally prepared tissue. (G–H) A group of neoplastic cells (arrow) invading the parietal peritoneum. Low- (J and L) and high- (K and M) magnification projection images of collagen architecture from normal peritoneum (J and K) and around tumor regions (L and M). (N and O) Representative Fourier transforms from the full images shown in K and M, respectively. Gaussian ellipses (blue) are fit to determine the collagen fibril asymmetry. MPM images were acquired using 780-nm excitation. Blue and yellow pseudocolors represent 355 to 425 nm and 450 to 550 nm emissions, respectively. For clarity, the SHG images are reproduced in gray in C, F, and I to M. Scale bars: 30 µm (A–C), 22 µm (D–F), 45 µm (G–I, J, and L), and 12 µm (K and M).
Figure 2
Figure 2
Standard histology and MPM/SHG of human ovarian epithelium and poorly differentiated adenocarcinoma. The same tissue specimens were visualized after formalin fixation and paraffin embedding followed by H&E staining (A and C) and before fixation using MPM/SHG of intrinsic tissue emissions (B and D). (A, B) Morphologically normal ovarian epithelium in the invaginations (arrow) and simple cysts (arrowhead) near the surface of the ovary. (C–D) Atypical neoplastic cells forming glandular structure (arrow) within desmoplastic stroma (arrowhead). Note presence of collagen fibrils imaged via second harmonic generation (SHG; blue), a feature not readily evident in conventionally prepared tissue. MPM/SHG images are acquired as in Figure 1. Scale bar, 30 µm in all images.
Figure 3
Figure 3
Ovarian surface epithelium carcinogenesis associated with conditional inactivation of p53 and Rb genes. Multilayered atypical cells forming papillary structures (arrows) can be identified by conventional microscopy (A) and MPM/SHG (B). Thickening of the collagen fibrils (arrowheads) near and at the basement membrane of early dysplastic lesion (arrow) is evident in the MPM/SHG image (C) with the SHG reproduced in grayscale for clarity (D). (E) Expression of eGFP in the OSE of a mouse ovary after a single ovarian intrabursal injection with 5 x 107 pfu/µl of AdCMVEGFP. Arrow indicates OSE; CL, corpus luteum; F, follicle; V, vessel. Green pseudocolor is GFP fluorescence and yellow is intrinsic fluorescence (<500 nm emission). (F) Morphometric evaluation of intrinsic emission images of mouse OSE. Size of individual cells and their nuclei and cytoplasm was assessed by estimation of area (white rings, µm2, mean ± SE). (G) Results are plotted for images collected at 8 days after intrabursal administration of AdCMVCre (Cre) or AdCMVLacZ (LacZ) to p53floxPRb1floxP mice. Unpaired t test yielded 2-tailed P = .0002 for whole-cell Cre versus LacZ (171.4 ± 12.4, n = 45 vs 120.4 ± 6.0, n = 53), P = .0466 for nucleus Cre versus LacZ (74.8 ± 5.3, n = 45 vs 53.0 ± 10.7, n = 26), and P = .0009 for cytoplasm Cre versus LacZ (96.58 ± 8.7, n = 45 vs 65.6 ± 3.78, n = 53). (H) At 34 days after transformation, unpaired t test yielded 2-tailed P = .0007 for whole-cell Cre versus LacZ (130.4 ± 6.5, n = 40 vs 92.1 ± 8.8, n = 36), P = .015 for nucleus Cre versus LacZ (70.0 ± 4.5, n = 40 vs 53.4 ± 4.9, n = 36), and P = .0007 for cytoplasm Cre versus LacZ (60.38 ± 4.3, n = 40 vs 38.6 ± 4.3, n = 36). All experiments have been performed in duplicates, and yielded similar results. Scale bars, 15 µm (A, B), 20 µm (C, D), 30 µm (E), and 12 µm (F).
Figure 4
Figure 4
Emission color changes in intrinsic fluorescence of EOC. Normal OSE (A and B) and neoplastic cells from the disseminated peritoneal EOC model (C and D) were analyzed for the relative amounts of LW/SW (510–650 nm/410–490 nm, green/red pseudocolor) intrinsic fluorescence. For reference, SHG from the collagen of the ovarian bursa is shown in blue pseudocolor. Intrinsic fluorescence images of OSE are manually marked (green masked areas, A and C). The white boxes mark the zoomed regions shown in B and D, respectively. (E) Analysis of average pixel intensities in the green masked areas shows that tumor intrinsic fluorescence is red-shifted with respect to normal OSE owing to a distinct redemitting cell population (arrows, D). Average LW/SW ratios are 1.00 ± 0.03 (mean ± SE, n = 9; RU, relative units) for normal OSE and 1.3 ± 0.1 (P = .0064, n = 7) for neoplasms. Data were acquired from 450 images at 780-nm excitation. Scale bars, 50 µm (A, C) and 25 µm (B, D).
Figure 5
Figure 5
Discriminating normal from neoplastic OSE in live mice using MPM/SHG. Images of normal (A–C) and transformed (D–F) OSE mice from the conditional inactivation of p53 and Rb model are acquired by exteriorizing the ovary and imaging with a standard objective lens. In zoomed images from A and D, respectively, OSE (arrows) are resolvable as a single layer in normal ovaries (B) and in multiple layers in neoplasia (E). In addition, the SHG channel (C, F) shows that the collagenous layer underneath of the OSE is visibly thicker in the neoplasia (arrowhead). OB marks the ovarian bursa and >> marks horizontal motion artifacts due to mouse breathing. Scale bars, 30 µm (A, D) and 15 µm (B, C, E, and F).
Figure 6
Figure 6
Intravital MPM/SHG imaging in the abdominal cavity using a microprobe objective lens (diagram, A). Normal peritoneum (B and C) and neoplastic cells of the disseminated peritoneal EOC model (D–G) imaged in vivo and in situ through an abdominal incision. Normal mesothelium is very thin, has low intrinsic fluorescence, and is essentially invisible using intravital MPM. However, neoplastic lesions are clearly visible at the invasive edge (D and E) and at the tumor surface (F and G). For clarity, SHG images are reproduced in grayscale in C, E, and G. Motion artifacts due to mouse breathing are marked with >>. The yellow meniscus at the bottom of B is the edge of field of view of the stick objective lens. Scale bar, 30 µm.
Figure 7
Figure 7
Imaging cathepsin activity using MPM/SHG. Images of the normal peritoneum (A, C, E, and G) and peritoneal tumors (B, D, F and H, arrows) from the disseminated peritoneal model after administration of ProSense 680 (red). Detection of neoplastic lesion using standard stereofluoroscopy (A and B) and subsequent mapping its tumor boundaries (D, arrowheads) in three dimensions with MPM/SHG. In this case, normal (C) and tumor (D) images were collected with a low-magnification objective (4x/0.28 NA) and displayed as projections, each from 100 images at 5-µm intervals. Discrimination between normal abdominal wall (E) and neoplasm (F) at cellular resolution with the stick objective. High-resolution MPM/SHG images demonstrate cathepsin activity in certain stromal cells (G, arrowheads) and neoplastic cells (H, arrow). In addition, they show alterations in the size and shape of neoplastic cells and tissue architecture as well as the orientation of collagen fibrils toward tumor boundary (H, arrowhead). Scale bars, 1000 µm (A, B), 500 µm (C, D), 80 µm (E, F), and 100 µm (G, H).

Similar articles

See all similar articles

Cited by 38 articles

See all "Cited by" articles

LinkOut - more resources

Feedback