The study found that axonal transport defects activate the PI3K pathway in an attempt to reduce oxidative stress and neuronal cell death in neurodegenerative disease models. Expressing constitutively active PI3K decreased cell death caused by polyQ repeats and Paraquat exposure but did not affect axonal transport blockages. Dominant negative PI3K disrupted normal huntingtin motility, indicating PI3K acts downstream of transport. Motor protein mutations and disease models showed increased levels of p-GSK3β, a PI3K effector, suggesting transport defects trigger the protective PI3K response.

1. www.buffalo.edu
Activation of the PI3K Pathway during Axonal Transport Defects can lead to Oxidative Stress-Induced
Neurodegeneration
Claire Thant, Megan Lamb, Timothy Hansen, Shermali Gunawardena
University at Buffalo Department of Biological Sciences
Summary
High levels of oxidative stress can be detected in neurons affected
by neurodegenerative diseases such as Parkinson’s (PD),
Huntington’s (HD), and Alzheimer’s diseases (AD). In addition to
oxidative stress, axonal transport defects and neuronal cell death
are also seen in these diseases. Here, we test the hypothesis that
axonal transport defects instigates oxidative stress causing
neuronal cell death. We found that Paraquat (a known inducer of
oxidative stress) ingested larvae exhibits axonal blocks and
neuronal cell death. Interestingly, expression of active
phosphatidylinositol 3-kinase (PI3K) (a kinase in the pro cell
survival pathway) suppresses Paraquat-mediated cell death but
not axonal blocks. Expression of active PI3K suppresses neuronal
cell death induced by expansion of polyQ repeats, but does not
affect axonal transport defects indicating that the PI3K pathway is
downstream of axonal transport defects. Additionally, dominant
negative PI3K disrupts the normal motility of HTT suggesting that
the PI3K pathway is directly linked to axonal transport. Intriguingly,
proteins in the PI3K pathway show functional interactions with
motor proteins and increased levels of glycogen synthase kinase
3 (GSK3 ), a downstream effector of PI3K, is observed in larvaeβ β
expressing expanded amounts of polyQ repeats and in motor
protein mutations. Taken together these observations suggest that
axonal transport defects likely activates the PI3K pathway to
decrease oxidative stress induced neuronal cell death and
degeneration.
References
Arvind K. Shukla, Prakash Pragya, Hitesh S. Chaouhan, D.K. Patel, M.Z.
Abdin, Debapratim Kar Chowdhuri, “A mutation in Drosophila methuselah
resists paraquat induced Parkinson-like phenotypes.” Neurobiology of Aging,
Volume 35, Issue 10, October 2014, Pages 2419.e1-2419.e16
Dolma K, Iacobucci GJ, Zheng KH, Shandilya J, Toska E, White JA 2nd, Spina
E, Gunawardena S. (2013) Presenilin influences Glycogen Synthase Kinase-
3beta (GSK-3 ) for kinesin-1 and dynein function during axonal transport. Humβ
Mol Genet. 2013 Oct 8.
Gunawardena, S. and Goldstein, L.S.B. (2001).
"Disruption of axonal transport and neuronal viability by amyloid precursor protein muta
Neuron 32:389-401.
Gunawardena, S., Her, L., Laymon, R.A., Brusch, R.G., Niesman, I.R.,
Sintasath, L., Bonini, N.M., and Goldstein, L.S.B. (2003) "Disruption of axonal
transport by loss of huntingtin or expression of poly Q protein in
Drosophila." Neuron 40:25-40.
Martindale, J.L., Holbrook, N.J. (2002) “Cellular response to oxidative stress:
Signaling for suicide and survival” J. Cel.. Physiol. 192: 1-15.
Acknowledgements
Special thanks to everyone at the Gunawardena Lab, as well as the UB
Center for Undergraduate Research and Creative Activities for funding
this project.
Figure
5.
p-GSK3β (S9)
Total GSK3β
Tubulin
ApplGal4
Roblk-/-
PI3K.CAAX
PI3K.21B
Htt128Q
Htt138Q
MJDQ77
MJDQ78
APPswe
p-Akt (S473)
Total Akt
A.
B. C.
Figure 5. Levels of p-GSK3 (S9) but not p-Akt(S473) is increased in bothβ
motor mutants and PolyQ disease genotypes.
A. Western blot analysis of a dynein motor mutant (Rob1k -/-), PolyQ and APP disease
genotypes (HTT128Q, HTT138Q, MJDQ77, MJDQ78, APPswe), and excess of PI3K
(PI3K.CAAX, PI3K21B) probed with antibodies against GSK3Beta (S9) which probes the
activation of the PI3K pathway, Total Akt, and Tubulin are also probed as a control. B-C.
Quantitative analysis reveals that levels of p-GSK3Beta (S9) are increased in both the
dynein motor mutant as well as the disease genotypes, while levels of p-Akt (S472)
remained unchanged. N = 1 gel.
Figure 6.
CSPA
WT KLC+/- Roblk +/-
PI3K92E PI3K92E;KLC PI3K92E,Roblk
Akt.Exel Akt.Exel;KLC Akt.Exel, Roblk
tor.WT tor.WT;KLC tor.WT,Roblk
B
C
D
E
F
14-3-3Zeta[07103] +/-
14-3-3Zeta[12BL]
+/-
14-3-
3Zeta[07103];
KLC
14-3-3Zeta[07103],
Roblk
14-3-3Zeta[12BL];
KLC
14-3-3Zeta[12BL],
Roblk
G.G
*
***
* **
**
*
ns
Figure 6. Proteins in the PI3K/Akt
signaling pathway genetically interact
with kinesin and dynein. A: Wild type
larval segmental nerves show smooth staining
(CSP). kinesin light chain (KLC +/-) and dynein
light chain (Roblk +/-) also show smooth
staining Bar = 50 m.μ B. Larvae expressing
PI3K92E.CAAX with 50% reduction in kinesin
or dynein show axonal blocks (Arrows). C.
Larvae expressing Akt show smooth staining,
while larvae expressing Akt with either a 50%
reduction OF kinesin or dynein show axonal
blocks (Arrows). D. Larvae expressing Tor
show smooth staining, while larvae expressing
Tor with either a 50% reduction in kinesin or
dynein show axonal blocks (Arrows). E-F.
loss of function OF 14-3-3Zeta (12BL), or
partial loss of function (07103) show smooth
staining in segmental nerves. When combined
with 50% reduction in either kinesin or dynein
both show axonal blocks (Arrows). G-H.
Quantitative analysis reveals that axonal
accumulations between the following
genotypes are significant as compared to the
Wild Type control: UAS-Akt.Exel;KLC +/-
(p=0.015), UAS-Akt.Exel, Roblk+/- (p=0.0004),
UASPI3K92E;KLC +/- (p=0.049), UASPI3K92E,
Roblk +/- (p=0.002), UAStorWT; KLC +/-
(p=0.007), UAStorWT, Roblk +/- (p=0.02), 14-
3-3Zeta[12BL] +/-, KLC +/- (p=0.0007), 14-3-
3Zeta[07103] +/-, KLC +/- (p=0.02). N = 5
larvae.
H.G
ns
ns
**
p=0.038
N=6 N=3
Htt138QmRFP TUNEL Merged
PI3K92E.CAAX;Htt138QmRFP
A B
Figure 3.
C
D E F
G.
Figure 3. Expression of active PI3K
suppresses neuronal cell death
induced by expression of expansion
of polyQ repeats A-C. Expression of
HTT138QmRFP causes neuronal cell death
as measured by the TUNEL assay. D-F.
Larvae expressing active PI3K
(PI3K92E.CAAX) with HTT138QmRFP
decreases the amount of neuronal cell death
G. Quantitative analysis reveals that the
amount of cell death seen in
PI3K.CAAX;HTT138QmRFP larvae are
significantly less compared to larvae
expressing HTT138QmRFP alone (p =
0.038.)Figure 4.
HTT15Q
HTT15Q
CSP
CSP
Overlay
CSP
HTT15QmRFP
HTT15Q-mRFP: PI3K.DNHTT15Q-mRFP
Figure 4. Dominant
negative PI3K disrupts
the normal motility of
HTT within axons
A. Expression of
HTT15QmRFP normally
shows smooth CSP
staining within larval axons
similar to Wild type larvae,
Note that HTT also is
smooth within axons B.
Expression of the
dominant negative PI3K, or
PI3K.DN with
HTT15QmRFP causes
CSP and HTT blockages.
PI3K/Akt signaling is overactive in
motor mutants as well as numerous
neurodegenerative disease
genotypes.
Proteins in the PI3K pathway show
functional interactions with motor
proteins.
Expression of excess polyQ repeats
causes axonal transport defects and
cell death.
Expressing constitutively active PI3K
protein is able to rescue HTT138Q
induced neuronal cell death, but not
axonal transport defects.
Paraquat ingestion causes axonal
transport defects and cell death.
Expressing constitutively active PI3K
protein with Paraquat ingestion has no
effect on axonal defects but decreases
neuronal cell death.
Expression of dominant negative
P13K causes axonal transport defects.
PI3k acts downstream of axonal
transport.
P13K pathway is likely activated due
to axonal transport defects and an
early oxidative stress response.
PI3K.CAAX;Htt138QmRFP
B
Htt138QmRFP
AHtt138Q CSP Merged
Figure 2.
Figure 2: . Expression of active PI3K does not affect axonal transport
defects induced by expression of expansion of polyQ repeats. A.
Expression of HTT138QmRFP causes accumulations of mutant huntingtin and
cysteine string protein (CSP) (arrows). Note that accumulations of CSP co-localize with
huntingtin (yellow dots, merged image.) B. Larvae expressing PI3K92E.CAAX with
HTT138QmRFP also contain accumulations of both mutant huntingtin and CSP.
Figure C-D. Quantified analysis reveals that the number of are not significantly
different between larvae expressing HTT138QmRFP, and larvae expressing both
HTT138QmRFP and PI3K92E.CAAX indicating that active PI3K does not have an
effect on axonal transport defects and that the PI3K pathway is downstream of axonal
transport. N = 5 larvae.
C.
ns
D.
ns
Figure 1. Ingestion of Paraquat causes axonal transport defects and neuronal
cell death. Expression of active PI3K suppresses Paraquat-mediated cell death
but not axonal transport defects . A, B. Wild Type (APPLGAL4) Drosophila larvae
raised on 0mM and 20mM Paraquat. Note axonal blocks in 20mM Paraquat. C, D, E.
Drosophila larvae expressing PI3K92E.CAAX raised on 0mM and 20 mM Paraquat. Note that
Paraquat-mediated axonal blocks are not rescued by excess P13K. (p = 0.4879.)
F, G. Wild Type (APPLGAL4) Drosophila larvae raised on 0mM and 20mM Paraquat show
statistically significant amount of neuronal cell death as assayed by the TUNEL assay. (p
=0.0165.) H, I, J. Expression of PI3K significantly suppresses Paraquat induced cell death.
(p = 0.0013).
Figure 1.
A.
C.
B.
D.
F G
H I
ns
ns
ApplGal4 0mM CSP ApplGal4 20mM CSP
PI3K92E.CAAX 0mM CSP PI3K92E.CAAX 20mM CSP
E
.
ApplGal4 0mM ApplGal4 20mM
PI3K92E.CAAX 20mMPI3K92E.CAAX 0mM
J.
p = 0.0165
*
p = 0.0013
**
N=9 N=10 N=6 N=10
N=12 N=5 N=12 N=10
A. B.
Paraquat &
PolyQ Expansion
Conclusions