oringham's blog

The original and replicate figures created of Calliandra haematocephala possess qualitative differences despite having the same methods followed in order to create them. Figure 1A and 2A are photos of different blooms on the tree, resulting in a large visual difference in the figure. Figures 1B and 2B are of the trunk and branches of the tree, and appear to be the most alike between all three components of the figure with only very slight differences in branches captured. Figures 1C and 2C also contain distinct differences. The borderlines for the states within the United States of America are outlined, and Florida is filled in red in figure 1C. Figure 2C does not possess borderlines for the states of the United States, and the state of Florida and the United States is a gray color. Additionally, different fonts were used to label each figure.

Overall, this laboratory exercise demonstrated the major elements that effect imaging in fluorescence microscopy. The net local concentration and degree of overlap of fluorophores and amount of the fluorescent dying in certain areas greatly effects the ability to achieve a bright and clear fluorescent microscopic image. Additionally, important microscope parameters such as the neutral density filters, and shutter/exposure time of the sample can greatly affect the brightness of an image and the rate at which fluorescent light decays over time, which is important to control in order to uphold the integrity of a sample.  

The rhodamine labeled f-actin did not appear in the composite image of the three different fluorophores due to the intensity of the fluorescence being too low to register. This could be due to photobleaching that had previously occurred in this specific area of the sample, or if the sample was not labeled adequately with enough fluorescent dye. Time-lapse data for all three fluorophores under the same condition revealed discrepancies in rate of decay and initial intensity for each fluorophore. A relative high initial intensity for DAPI labeled dsDNA can be explained by the relative high net local concentration of bright fluorophores. Each nucleus contains a high concentration of dsDNA, which when stained with DAPI, creates a large solid fluorescent region with overlapping fluorophores. This differs from both tubulin and f-actin, which are of tube like nature, and appear as porous regions of interest where background light can seep through and be analyzed, making the initial brightness in the region of interest inherently darker. This difference is also reflected in the rates of decay, where in samples where there is less light to be diminished, the rate of decay appears much slower as it approaches the plateau of photobleached darkness. Time-lapse data for fluorescein under stained tubules under different conditions demonstrated discrepancies in rate of decay and initial intensity for all three conditions. The relative low rate of decay and low initial intensity for the images taken with a neutral density filters and auto shutter can be explained by the effect of the neutral density filter. The collective effect of the neutral density filters decreases the intensity of the epi-illumination light path by a factor of 32. This in turn lowers the ability for the fluorophores to be excited, causing a low initial intensity and low excitation, which leads to a lower rate of decay. The auto shutter condition displays a lower rate of decay compared to the open shutter condition due to the lower exposure to light over time. When the shutter is left on automatic, the sample is only exposed to light during the time an image is snapped as opposed to the entire duration of the time lapse. This lessens the time window that fluorophores can covalently bond to oxygen and other elements, and less brightness is lost over time. The open shutter condition leaves the sample exposed to the epi-illumination light path during the entirety of the time-lapse, allowing more time for covalent bonds to form with fluorophores and thus lose more brightness intensity over time. This explains the high relative rate of decay.  

Fluorescein labeled tubulin was then time-lapse imaged under several conditions: auto shutter, open shutter, and open shutter with neutral density filters. Data was analyzed graphically for all three conditions (Figure 3). Open shutter data was shown to have the highest initial intensity value and rate of decay (Table 2). Auto shutter data contained midrange values for both initial intensity and rate of decay (Table 2). Time lapse data with neutral density filters and an open shutter demonstrated the lowest values for initial intensity and rate of decay (Table 2). Selected time lapse images visually demonstrate the difference in excitation intensity over time during the photobleaching process under all three conditions (Figure 4).

Photobleaching of LLC-Pk1 pig kidney epithelial cells was done with an open shutter and no neutral density filters for all three types of fluorophores. Time-lapse imaging data was collected and analyzed graphically for all three fluorophores (Figure 2). It is apparent that the DAPI labeled dsDNA has the highest initial intensity, and fastest rate of decay (Table 1). The rhodamine labeled f-actin holds both the lowest initial intensity value and rate of decay (Table 1). Fluorescein labeled tubulin have midrange values for both initial intensity and rate of decay (Table 1). 

Images of LLC-Pk1 pig kidney epithelial cells were taken under three different fluorescence filters in order to capture cellular structures stained with different fluorophores in the same region (Figure 1). These images were then superimposed on one another to create a comprehensive display of three cellular elements; dsDNA stained with DAPI (blue), tubulin stained with fluorescein (green), and f-actin stained with rhodamine (red). DAPI and fluorescein fluorophores showed visible fluorescence intensity and were captured in the image. The rhodamine stained f-actin did not appear in the composite image. Reasons for this disappearance is detailed in the Discussion.

Photobleaching in standard microscopy results when a fluorophore in its excited state covalently bonds with another molecule and renders it unable to absorb or emit light, thus darkening the overall image. The rate at which photobleaching occurs is a function of excitation intensity and time and can differ based on certain variables. This lab explores how different conditions of the microscope and fluorescent elements of the sample effect the rate of photobleaching.

Figure 1: A young felis catus of the tabby breed relaxing. (https://flic.kr/p/uwcAuk)

            Using a dynamic approach to modeling disease allows for a vast amount of advantages when addressing public health problems. Many results of mapping and modeling a progressive disease demonstrate epidemiological aspects that can be useful in discovering why the disease came to be and how it could be treated. Additionally, models can demonstrate the efficacy of routine screening and other forms of preventative care for those who are more at risk of certain diseases and conditions. Identifying important forms of upstream prevention can drastically reduce the spread and severity of disease. Modeling can also be useful in discerning the relationships between multiple interacting diseases, shedding light on feedback loops that amplify symptoms and chronic conditions. Disease modeling can also be used on a single patient level, where certain input signs and symptoms can lead to an output that would not have been reached by traditional means. Overall, the benefits of using simulation dynamics for modeling disease are immense, and can steer public health officials and medical practitioners to a sound and logical answer to questions that seem to have none.

There are many different types of modeling systems that are aimed at modeling different types of situations and observing different types of outcomes. Some are more simple than others, depending on the complexity of the system and the problem at hand. Simplest of all is the decision tree. Decision trees provide a logical and linear structure for a decision made and the event that occurs due to this decision. An example with respect to public health is a simple and easily distinguished outcome; a patient receives treatment and the outcome is they live, they do not receive treatment and they die. This model is very useful when dealing with situations in which there are a limited number of possible events, and there is a short timeline for the scenario. However, decision trees are not as useful when dealing with high variable and complex situations. Markov Models are slightly more complex simulations, which model chains of events in a “memoryless” manner, where predictions are only based on current states with no regard to past states. These models can be used to “many features present in the clinical process, such as risk of disease over time, [and] changing health states over time” (Kuntz et al. 2013). However, its disregard for past events leaves many important variables out of the equation that greatly impact future events. A final and more complex system of modeling is dynamic models, also known as infectious disease models. This type of model typically involves the simulation of interactions between humans and humans with other species. In order to simulate these interplays, complex differential equations must be used, leading to a more complex model. The transmission of disease based on spatial details and individual interactions can be be greatly detailed by this modeling system, leading to accurate quantification of impact of the disease. This data can allow simulation of interventions as well, so that transmission can be stopped. All of these modeling systems can be useful to predict and display public health issues and interventions.