DC: Conceptualization, Investigation, Writing - original draft, Writing - review & editing, Data curation, Formal analysis
LK: Data curation, Formal analysis, Methodology, Writing - original draft, Writing - review & editing, Investigation
SG: Investigation, Data curation, Formal analysis
JM: Data curation, Formal analysis, Writing - review & editing, Investigation, Supervision, Funding acquisition
A circulating tumor cell (CTC) is a type of cell that is shed from solid tumors, swept away in the bloodstream or lymphatic system, and has the potential to cause tumorigenesis at a secondary location. Here we describe an early pupal leg system to study CTCs
Schematic displaying developing
Although they were discovered in the 19th century (Wang
Despite their importance, CTCs are rare in the blood (approximately 1 CTC per 10
9
blood cells) (Alix-Panabières and Pantel 2014), presenting an obstacle to those isolating and studying them (Pantel 2016). Due to this technical limitation, most work thus far has concentrated on studying CTCs
The first step to answering this question is to develop
To find a
Analysis of CTCs in three time-lapse movies quantified three cellular parameters: quantity of CTCs, size (measured using 2-dimensional surface area), and shape (measured using circularity). To do so, we genetically induced tumors in the larval stage and studied CTCs in the developing pupal legs. We consistently obtained GFP-labelled tumors located in the developing legs, as well as many GFP-labelled individual and groups of cells moving around them, which resemble CTCs observed in previous studies (Aceto
The size of the CTCs also varied between all three movies (Fig. 1G; Mov. 1-3). Our statistical analysis confirmed that the area datasets from each movie were significantly different from one another (Mann-Whitney U p-value is < 2.2
-16
. Kruskal Wallis p-values are as follows: Mov.1 and Mov.2: 0.01846, Mov.1 and Mov.3: <2.2
-16
, Mov.2 and Mov.3: <2.2
-16
). Despite these differences, 81% of the CTCs in all the movies were smaller than 100 µm
2
(Fig. 1G). These smaller CTCs have a far lower variation in size than the larger CTCs (<100 µm
2
Stdev = 21.8 = vs >100 µm
2
Stdev = 146.3). We also used the software
Our
The difference in the CTC numbers and size needs to be studied in more detail. Although the videos analyzed belong to flies of a similar genetic strain and the flies were exposed to similar environmental conditions, future studies are needed to test whether this variation is a property of the system or not. Some of the variation observed among the three movies can be partially due to the difference in size and locations of tumors generated by the MARCM clones. Our current working hypothesis is that as the peripodial membrane is broken during development, the higher the number of tumors (
In addition, we found a much higher percentage of CTCs cluster (37%) compared to previous work (2.6%) (Aceto
To conclude, this work shows that using live imaging methods and multiple fluorescent markers,
Live imaging of the early pupal stages was performed between the rupture of the peripodial membrane and before head eversion. For live imaging, pupae were mounted in halocarbon oil (series 700; Halocarbon Products) on a coverslip (Sigma) and imaged with a laser 510 scanning confocal microscope (ZEISS model) at 25 degrees with a 40× objective, using LSM Browser software (ZEISS). z-stacks had a 1μm step size.
Once the time-lapses were recorded, the computer program
To collect information about the quantity of CTCs and CTC clusters, cells were outlined in each frame of the videos. Thus, the count of CTCs obtained from this analysis does not reflect how many CTCs were present in the movies, but rather, how many were present in each individual frame of the movies. The size (area) of CTCs was measured in both individual CTCs and CTC clusters, and both types of CTCs were grouped in the same category for our analysis (Zhou
Tumorigenic cells in the fruit fly pupae were generated by using tumor suppressor genes
|
|
|
|
||
hs-flp22; ubi-Cad::GFP, UAS-mRFP; act<y+<Gal4 |
Levayer
|
|
P{ry[+t7.2]=hsFLP}22, y[1] w[*]; P{ry[+t7.2]=neoFRT}82B bon[21B]/TM3, Sb[1] |
Bloomington stock center |
43660 |
UAS-GFP.S65T; FRT82B,tub-Gal80/FRT82B |
Wu
|
|
|
||
UAS-GFP.S65T/UAS-RasV12; FRT82B,tub-Gal80/FRT82B,scrib1 |
Wu
|
|
P{ry[+t7.2]=hsFLP}22, y[1] w[*]; P{ry[+t7.2]=neoFRT}82B bon[21B]/TM3, Sb[1] |
Bloomington stock center |
43660 |
|
||
yw, eyFlp; Act>y+>Gal4, UAS-GFP; Tub-Gal80, FRT79E, |
Dunn
|
|
UAS-RasV12, M6 -/- , FRT79E/TM6B |
We thank Canadian Mennonite University for funding and support. In addition, we are also very grateful to Ellen Larsen, Hillary Jorgenson, anonymous reviewers, and editors that provided valuable comments to improve the manuscript. Stocks obtained from the Bloomington Drosophila Stock Center (stock number 43660) were used in this study.
This work was supported by Canadian Mennonite University.