Team:Berkeley/Results

From 2010.igem.org

General Experimental Set Up

Our payload delivery assay involves the following steps:

1. Feed the payload bacteria to the choanos and induce self-lysis with arabinose. Induction of self lysis and feeding are not necessarily at the same time.
2. Use fluorescent microscopy to detect delivery of our GFP payload.
   

Challenges

One of the challenges we faced in assaying for payload delivery was finding a media that the Choanoflagellates liked and that accomodated lysis. After assaying several different media formulations, including CGM, LB, TB, ASW and four different mixtures of ASW and LB, we found that CGM media was the best media for Choano viability, E. Coli lysis, and protein delivery. For more details, see our characterization of the self-lysis device.

The major challenge we faced when assaying for payload delivery was determining when to induce self-lysis and when to look at the choanos under the microscope. For induction of self lysis, we tested timepoints from 90 minutes to 30 minutes after feeding the choanos (-90 to +30 minutes), with 15 minute intervals. For looking at the choanos under the microscope, we tested timepoints from 1 hour after feeding to 24 hours after feeding.

From our experiments, we found that induction at the same time as feeding, or 15 or 30 minutes after feeding yielded the most delivery events. For timepoints before -15 minute, delivery events were undetectable or happened extremely infrequently.

We also observed that choanos seem to eat and digest bacteria with 1-2 hours after feeding, and that the best time to look for successful delivery was between 3 and 5 hours after feeding.

Controls

Our first step before assaying was to establish controls for all of our assay experiments. Based on how our parts should work, and work done in the Anderson Lab in mammalian cells, we expected that a sucessful delivery event would consist of diffuse GFP fluorescence throughout a choano's cytoplasm. With this in mind, the controls we ran regarding the choanos were:

Payload cells with no payload delivery device

This control was to establish a baseline of what normal eating and digestion of bacteria in choanos looks like. We observed that choanos seem to eat with 3-5 minutes after feeding. During the first hour or so, one generally sees bacteria in food vesicles , either as rod shapes or circles (rod shaped bacteria 'head on'). They are very bright and have signifcant halos. Slowly, we end up seeing fluorescence in small circular objects.The fluorescent intensity is less than the initial bacteria in food vesicles, and we believe flourescent payload is in the food vacuole. Eventually, the fluorescence disappears altogther, degraded by the vacuole of the choano.

Choanos fed just payload cells with no payload delivery device after 1 hour. You can see rod shaped bacteria in food vesicles in the choanos. Imaged at 100 ms, green channel exposure

Choanos fed just payload cells with no payload delivery device after 3 hours . You can see a couple rod shaped bacteria inside of of choanos as well as many choanos with dimmer fluorescent circles, which we believe is gfp in the food vacuole, imaged at 300ms green channel exposure

Payload cells with uninduced payload delivery device

This control was to ascertain that any delivery events we got were from induction with arabinose, and not just a normal phenotype exhibited in when choanos eat cells with self-lysis, vesicle buster and payload.

Choanos fed just uninduced payload cells with payload delivery. We did not see any successful delivery and this image shows choanos with bacteria in their food vacuoles, in the process of normal digestion. Imaged at 300 ms, green channel exposure

Cells with payload and self lysis but no vesicle buster

This control was to confirm that the successful delivery phenotype we saw was due to self-lysis device and the vesicle buster, not just self-lysis alone. We wanted to make sure that the what we were seeing was payload that had escaped the food vesicle, not just payload in a food vesicle.

Choanos fed just payload cells with only self-lysis after 3 hours. You can see fluorescence inside the food vacuole of choanos. Imaged at 300 ms, green channel exposure

Cells with payload and vesicle buster, but no self-lysis

This control was once again to make sure that if we saw a successful delivery event, we could be certain that it was a phenotype unique to cells with induced self-lysis and vesicle buster.

Choanos fed just payload cells with no payload delivery device after 1 hour. You can see rod shaped bacteria in food vesicles in the choanos. Imaged at 100 ms, green channel exposure

Additionally, since all of these assays are microscopy based, we kept the settings on our microscopes constant and kept careful notes on exposure and magnification in order to avoid getting misleading results.

Payload Delivery

The following images are of successful payload delivery events. As the images of the control above showed, the phenotype of GFP throughout the cytoplasm shown below is unique to constructs with self-lysis and vesicle buster, and is consistent with what we'd expect to see based on the biology of our parts.

Successful delivery event, imaged at 100ms green channel exposure	Successful delivery event, imaged at 300ms green channel exposure
Confocal Imaging
Z-stack imaging with 1.5 micron slices Z-stack imaging	Ortho-view of z-stack imaging

Hemocytometry

We also used a hemocytometer to quantify the frequency of delivery. For an assay with a +15 minutes induction, the number of delivery events were 9/130, 5/57, and 9/179 (events/total choanos) or an average of 7% (±1% standard error).

Specific Protocols WILL FINISH TOMORROW or time points before feeding, we induce self lysis in a testube, and then put it in a 37 degree shaker until it is time to feed the choanos. For time points after feeding, we add arabinose to the choanos and keep them incubating at 37 C until we assay.