Return to search

Diagnosing Changes in Cells Using FTIR Microspectroscopy

Fourier transform infrared (FTIR) microscopy has shown promise as an analytical tool for detecting changes in cells and tissues, such as those due to viral infection, apoptosis induction or malignancy. In many cases, diagnosis via FTIR microscopy can be undertaken on a timescale shorter than that required for other physical or histological techniques.
In this work we have used FTIR microscopy to study Vero cells that have been infected with herpes simplex virus (type I) and adenovirus. We have studied cellular samples at various time intervals following exposure to the virus. Several spectral regions were identified that allow discrimination between infected and uninfected Vero cell samples at 24 hours post exposure to both HSV1 and adenovirus. Spectral features were also identified that could be used to discriminate infected cells within 2-6 hours after exposure to both viruses. FTIR microscopy is therefore a useful tool for following the kinetics of viral infection in the 2-24 hours time range, at least at the levels of infection used in this study.
In a second type of study, FTIR microscopy was used to study apoptosis induction in acute lymphoblastic leukemia T-cells. Apoptosis was induced in T-cells in three different ways. We show that FTIR microscopy can be used to distinguish T-cells in the early stages of apoptosis from normal cells. We also provide data that may suggest that FTIR microscopy can distinguish cells that have undergone apoptosis via different pathways.
For most of the FTIR microscopic studies on cellular samples we have focused on the collection of spectral data in the 1500-800 cm-1 region. Spectra were collected for control cells and variously treated cells. The two sets of cells were then analyzed statistically using: 1) pair-wise comparison, 2) logistic regression, 3) partial least square regression, 4) principle component fed linear discriminant analysis and 5) hierarchical cluster analysis. The statistical analyses rigorously quantify to what extent treated and untreated cells can be distinguished. Since different statistical methods give differing results for the same data, it is important the right statistical method should be applied. The basis for these differences is discussed.

Identiferoai:union.ndltd.org:GEORGIA/oai:digitalarchive.gsu.edu:phy_astr_diss-1046
Date13 May 2011
CreatorsGuo, Jing
PublisherDigital Archive @ GSU
Source SetsGeorgia State University
Detected LanguageEnglish
Typetext
Formatapplication/pdf
SourcePhysics and Astronomy Dissertations

Page generated in 0.0016 seconds