Retinal Stem Cell in Fish Model Systems
Dawn Sadler
April 26, 2005
According
to past research, cichlid fish have many promising results in locating retinal
stem cells. Specific cichlids that are being studied are Jack Dempsey and Red
Borleyi fish, both of which were previously unstudied species. The major goals
in this research are to see if these specific species of cichlids are suitable
models for retinal stem cell research first, then eventually determine when
cell division occurs more rapidly during light or dark hours in a day and if in
fact cell division can be controlled by a circadian clock. In reference to the
article, Daily Rhythm of Cell Proliferation In the Teleost Retina, data was used and the analysis was made.
This article studied the cichlids, Haplochromis burtoni. The results of using a parametric
t-test showed that cell division occurred more during dark hours then light
hours in a light-dark cycle.
Question:
Is
there a higher rate of cell division in the presence of light in comparison to
the absent of illumination?
Hypotheses:
Ho: µ Dark ² µ Light
HA:
µ Dark > µ Light
Sampling
Design:
The
H. burtoni fish used
to collect data were from wild stock that was caught and raised in an aquarium
similar to their natural environment. There were three experiments performed.
The second experiment was a replica of the first. The first and second
experiments were based on a light-dark cycle. The third experiment was based on
a cycle of constant darkness. The fish were moved from their natural environment
to an experimental aquarium. Two fish form the same tanks were sacrificed in 4
hours intervals starting at 11:00 (11:00, 15:00, 19:00, 23:00, 03:00 and 07:00)
for all three experiments.
After
sacrificing the fish, their eyes were removed and fixed using paraformaldehyde.
The retinas were then embedded in an agar and sectioned on the cryostat at 20µm
along the nasotemporal axis. The sections were re-hydrated and washed in HCL. A
normal goat serum was applied to block nonspecific binding of secondary
antibody. Primary anti-PCNA antibody was added and the tissue was incubated.
Cells labeled with PCNA (proliferating cell nuclear antigen) antibody were
visualized and counted using a light microscope. The number of labeled cells was expressed per unit of
'retinal length'. The retinal length is from the outer nuclear layer measured
from one germinal zone to another. Since the section thickness was constant,
the number of labeled cells expressed as a function of retinal length was
directly proportional to the volumetric density sampled from each retinal
section (J.F. Chiu et al.). The mean and standard deviation of the cell
densities for each retinal were calculated to be about 20 retinal sections from
two fish.
Data
Analysis:
Using
the format provided by the article, two similar graphs and data were
formulated. From the graphs in the article, an estimation of the number of
cells/mm at specific time intervals was made. These numbers were created to
imitate what values could have been accounted for.
Graph
1
This graph show the number of PCNA - labeled rod
precursor cells plotted as a function of time of day in a standard light-dark
cycle and under constant darkness. The white bars represent light hours and the
black bars represent the dark hours. It can be seen that there is a significant
difference of counted cells in the dark hours then during the light hours. In
this specific graph represents
experiment 3, constant darkness cycle.
|
11:00 AM |
15:00 PM |
19:00 PM |
23:00 PM |
3:00 AM |
7:00 AM |
11:00 AM |
15:00 PM |
19:00 PM |
23:00 PM |
3:00 AM |
|
32 |
39 |
35 |
70 |
73 |
52 |
8 |
20 |
20 |
18 |
46 |
|
36 |
31 |
35 |
74 |
75 |
56 |
10 |
26 |
18 |
12 |
39 |
|
36 |
30 |
37 |
69 |
81 |
59 |
11 |
17 |
21 |
13 |
40 |
|
30 |
37 |
36 |
75 |
83 |
52 |
7 |
23 |
18 |
12 |
41 |
|
34 |
34 |
35 |
76 |
81 |
51 |
6 |
25 |
19 |
19 |
43 |
|
36 |
33 |
30 |
72 |
79 |
50 |
5 |
27 |
21 |
16 |
45 |
|
34 |
34 |
30 |
73 |
62 |
50 |
2 |
20 |
18 |
13 |
47 |
|
30 |
29 |
32 |
70 |
79 |
42 |
15 |
28 |
20 |
9 |
49 |
|
29 |
32 |
32 |
74 |
85 |
61 |
3 |
23 |
21 |
16 |
43 |
|
29 |
36 |
29 |
75 |
81 |
59 |
7 |
29 |
21 |
17 |
43 |
|
28 |
29 |
30 |
70 |
89 |
61 |
9 |
20 |
22 |
18 |
42 |
|
32 |
30 |
30 |
69 |
89 |
48 |
8 |
30 |
19 |
12 |
40 |
|
34 |
34 |
30 |
71 |
87 |
50 |
8 |
32 |
20 |
11 |
41 |
|
27 |
36 |
31 |
73 |
88 |
53 |
10 |
25 |
20 |
19 |
40 |
|
35 |
32 |
34 |
75 |
82 |
56 |
11 |
23 |
21 |
15 |
42 |
|
35 |
30 |
29 |
70 |
76 |
58 |
19 |
24 |
20 |
19 |
41 |
|
39 |
38 |
31 |
75 |
79 |
52 |
24 |
27 |
19 |
12 |
43 |
|
31 |
29 |
32 |
73 |
79 |
50 |
15 |
29 |
21 |
14 |
46 |
|
30 |
37 |
34 |
72 |
80 |
51 |
20 |
29 |
22 |
21 |
47 |
|
32.47368 |
33.15789 |
32.21053 |
72.42105 |
80.42105 |
53.21053 |
10.42105 |
25.10526 |
20.05263 |
15.05263 |
43.05263 |
Table 1
These
numbers were created to imitate what values could have been accounted for. The
last rows are the averages of the numbers. The average is based off of the
graphs provided in the article. In the statictica program, Shapiro-Wilk'd test
was preformed to test for normal distribution for each column. In order for the data to be considered normally
distributed, the p-value ³ 0.05.
Statistical
Method
Once the created
data was formatted to read the counted cells in one column and the label of
"Light" or "Dark" in the second column this data was
imported in to statistca to create t-test to test for the difference of the number
of cells counted in the light hours to the cells counted in the dark hours.
20 Light 20 Light 21 Light 20 Light 19 Light 21 Light 22 Light 18 Dark 12 Dark 13 Dark 12 Dark
Summary
Statistics
|
Mean Light |
Mean Dark |
t-value |
SD Light |
SD Dark |
|
25.57018 |
52.83158 |
-11.3833 |
9.030890 |
33.58852 |
Conclusion
After
studying and analyzing the data, the results of the t-test show that the null
hypothesis, Ho: µ
Dark ² µ Light can be rejected. The number of cells counted
during dark hours in a light-dark cycle and constant darkness is significantly
greater than those cells counted during light hours.