~ This web page was produced as an assignment for an undergraduate course at Davidson College ~
“Mitochondrial
pathology and apoptotic muscle degeneration in Drosophila parkin
mutants”
Jessica C. Greene, Alexander J. Whitworth, Isabella Kuo, Laurie A. Andrews,
Mel B. Feany, and Leo J. Pallanck
(2003), PNAS 100 (7), 4078-4083.
Paper Summary:
Parkinson’s disease is a neurodegenerative disorder characterized
by a loss of dopaminergic neurons in the substantia nigra and accumulation of
Lewy bodies in humans. Although no known molecular mechanisms are known, several
experiments have indicated mitochondrial dysfunction as a major causative factor.
An important gene in the identification of factors causing Parkinson’s
is the parkin gene, where loss of function mutations cause an early
onset form of Parkinson’s known as autosomal recessive juvenile parkinsonism.
The researchers in this paper decided to investigate Parkinson’s disease
through Drosophila, by creating a parkin- mutant and comparing
symptoms and mechanisms to those thought to occur in humans. Parkin in humans
has recently been demonstrated to be an E3 ubiquitin protein ligase, interacting
with ubiquitin-activating (E1) and ubiquitin-conjugating (E2) enzymes.
Greene et al. began by finding the gene in the Drosophila
genome that matched the human Parkin sequence. They searched the Berkeley Drosophila
Genome Project Database with the Parkin protein sequence to query a six-way
translation. Only one gene was identified, and the cDNA was then sequenced and
found to have 42% amino acid identity and 59% similarity overall with human
Parkin. The cDNA sequence was also compared to the genomic DNA sequence to identify
splice junctions for introns and exons. The structure of the Drosophila
gene was also very similar to that of the human Parkin, bearing a ubiquitin-like
domain, ring-finger domains, and an IBR domain (Figure 1A).
The parkin mutants were generated using P element insertions:
genomic DNA was analyzed from approximately 5,500 flies with multiplex PCR using
a primer specific to the P element terminal sequence, and other primers corresponding
to sequences in the parkin gene. One line was recovered, parkEP(3)LA1,
which had an insertion 71 bp upstream of the parkin start codon (Figure
1C). Imprecise excision alleles were then created by mobilizing the transposon
and favoring coincident deletions, which were then determined through sequencing.
Several alleles were identified as having the entire parkin sequence
deleted, thus becoming null alleles, parkin13 and parkin45.
Another mutation, parkin25 had most of hte parkin gene
deleted but left the latter portion intact. A chromosome was also revealed
to have precisely deleted parkEP(3)LA1, which was then used as a control
chromosome (designated parkrvA). The sequence of the gene was then
altered to include a polymorphism at codon 240 to match the parkin
amino acid sequence predicted from the Berkeley Drosophila Genome Project.
PCR was conducted to introduce sequence changes designed to improve translation,
and to introduce restriction sites. Then this new construct was sequenced to
ensure the integrity of the parkin coding sequence. A Northern Blot
analysis was then performed with a parkin probe to ensure the construct
was being made and was still identifiable with a parkin sequence probe
(Figure 1B).
The flies produced with the parkin mutations were then
analyzed for the behavioral differences compared to control, wild-type flies.
Flight tests were done by dispensing 20 flies through a funnel into a graduated
cylinder that had a sheet covered in vacuum grease on the sides. The flies stuck
to the sheet where they landed, the sheet was then removed, and the flies were
counted in each region of the five regions on the sheet. Climbing assays were
done using a countercurrent apparatus, and 20-30 flies were placed into a chamber
then given 30 seconds to climb 10 cm. If flies completed this task successfully,
they were moved to another chamber and this process was repeated 5 times. After
five trials, the number of flies in each container was counted and the climbing
index was calculated the same way as the flight index: the weighted average
of the region or chamber where the flies had ended up, divided by four times
the number of flies in the assay (Figure 3).
The results of these experiments showed that the parkin
transcripts are being made at all developmental stages, from embryos to adults,
with the authors noting a particularly high abundance in adults (Figure 1B).
parkin null flies showed developmental delay, significantly reduced
longevity, male sterility, and severe defects in both flight and climbing ability.
Each of these symptoms was analyzed for a cause, hoping to be linked to the
human form of AR-JP (autosomal recessive juvenile parkinsonism). The male testes
were analyzed and it was discovered that the sterility was a result of a late
defect in spermatogenesis, which proceeded normally until the 64-cell germ-line
cyst that normally separates into mature sperm cells failed to divide. Structural
analysis of developing spermatids showed structural irregularities in the sperm
tails, as well. The Nebenkern integrity, a specialized mitochondrial derivative,
was severely disrupted with varying degrees: some showed reduced Nebenkern,
others showed multiples (Figure 2).
The flight problems were a result of a partially penetrant downturned
wing phenotype, with the penetrance increasing with age. Climbing ability decreased
at the same rate as control flies, however the parkin mutants performed
consistently lower, suggesting they being adult life with a reduced ability.
To address these issues, the researchers used a UAS/GAL4 system to express parkin
in defined tissues. It was discovered that parkin expression in the
mesoderm rescued wing posture, flight, and climbing phenotypes, demonstrating
that parkin function is required in the musculature.
Histological analysis revealed severe disruption of the muscle
integrity, and another rescue was performed by expressing parkin ectopically
in muscles. However, the muscles involved in the jump response of the fly and
the larval body-wall muscles were morphologically and functionally normal in
parkin mutants. Thus, only a subset of muscles are affected by the
loss of parkin. The disruption of muscular integrity was shown to be
a result of a decrease in density of myofibrils, a broadening of the myofibril
Z-line, and a shortening of the sarcomere length, but the most noticeable difference
from the control flies were the swollen mitochondria with severe disruption
and disintegration of the cristae. Transgenic expression of parkin
in the musculature restored the myofibril integrity and mitochondrial morphology
(Figure 4).
To determine if the muscle degeneration was being caused by apoptotic
cell death, the indirect flight muscles were subjected to TUNEL staining. After
96 and 120 h of puparium formation, no TUNEL staining was detected in parkin
mutants or control flies but a dramatic increase was observed in 1 day old adult
parkin mutants compared to the control adults of the same age (Figure
5). The researchers concluded that this showed the mitochondrial defects resulted
in cell death through an apoptotic mechanism.
Experiments were also performed to check the neuropil integrity,
which revealed appropriate development of the major brain centers and no neuronal
loss was observed (Figure 6). The only difference between control and parkin-
flies was some shrinkage of the dorsomedial dopaminergic cell cluster cell body,
as well as decreased staining of the proximal dendrites.
Critiques:
The first problem I see comes in the introduction, where the authors
state “The finding that Parkin functions as a ubiquitin protein ligase
indicates that failure to label specific cellular targets with ubiquitin is
responsible for dopaminergic neuron loss in AR-JP”. While this statement
may be true, there is no data supporting data, nor any other papers quoted or
referenced showing that research has been conducted to lead to this statement,
and this seems an unlikely conclusion with no experiments behind it. While they
then go on to say that experiments premised on this hypothesis have led to several
potential cellular targets of Parkin, there is no evidence supporting the view
that it must be lack of specific ubiquitin labeling that causes the neuron loss.
The second critique is that the cDNA used apparently already had
a mutation when comparing the cDNA sequence to the genomic sequence, because
the researchers had to alter the polymorphism (a mutation that does not change
the biological function significantly) at codon 240. Also, through PCR, many
different things were tagged on to the parkin coding sequence, and
this may have changed the structure of the protein enough to cause a slight
alteration in the function. While they did a Northern Blot test to show that
the mRNA was being produced from this gene in the transgenic flies, this does
not show that the parkin protein was acting in accordance with its
normal function.
With the assays testing climbing abilities, the test they used
was an experiment originally designed for phototaxis experiments, not climbing
experiments for flies, and they also modified the experiments used for testing
flight ability. While they did explain how they performed the tests, there is
no way for us to know that these tests are accurate measurements of ability.
In the results section, parkrvA was described as the control
chromosome because the mobilized transposon had deleted itself, however I wonder
why they did not simply use a chromosome that had not undergone the transposon
deletion at all to ensure a completely normal wild-type control.
In Figure 1B, the Northern Blot Analysis, the lanes appear smeary
or out of focus, and while the authors state that the adult fly has a significantly
higher amount of parkin being produced, there is no control showing
how much RNA was loaded into each lane, thus we do not know if this is an accurate
representation of the concentration in various developmental stages. Also in
the figure legend the researchers mention a band in adults that is larger than
the 1.7 kb parkin transcript, yet this extra band is not clarified
or even mentioned elsewhere in the text.
In Figure 1C, it would have been helpful to give more accurate
measurements of the deletions in the Parkin gene. While there is a schematic
drawing, they do not specify which amino acids have been deleted and it is difficult
to understand the parkZ472 and parkZ4437 constructs. The constructs
are only explained in the diagram and nowhere in the text, thus when they are
used again in later images, it is hard to tell what the experiment was testing
when using the constructs.
Figure 2 is a well-done experiment showing both control parkin+
as well as parkin-, and using arrows and arrowheads to point out the
important parts of the diagram so that the reader may understand even if they
are not familiar with the structures being shown, and there is quite obviously
a difference between the controls and mutants, supporting their hypothesis.
In Figure 3 E and F, it is interesting to note that Rescue 1 had
a result much closer to that of the wild-type than Rescue 2, and in the case
of the climbing index, Rescue 1 restores climbing to a higher ability than the
control itself shows. This difference may have affected their later experiment
showing the rescue of the parkin- phenotype by transgenic expression
of parkin in Figure 4, because the rescue the experimenters used was
Rescue 2. In Figure 4C, the authors note that some vacuoles are still present
in the muscle that has been rescued, however perhaps if they had used Rescue
1, this would not have been the case.
A critical problem in the argument of the researchers comes at
Figure 6, when their image shows that loss of parkin does not cause
general neuronal degeneration of dopaminergic neuron loss, such as that in AR-JP.
This is only a problem for later, when they try to show that the parkin-
pathway mechanisms in Drosophila are applicable to human AR-JP, because
if the flies are not exhibiting a key part of the human disease, how likely
are the other connections that have been made with the mitochondrial degeneration.
For the most part, this paper displays their data well, showing
the important images with clarity, yet there are a few sections that say data
not shown where it might have been better to show at least a few images supporting
their statements in the text.
Future Experiments:
Since mitochondrial degeneration and cell death appear to be the
main focus of this paper, perhaps a future experiment could leave parkin
intact and induce mutations in other genes that affect mitochondrial formation,
to see if their loss-of-function also cause similar problems to parkin
null alleles. It would also be an interesting experiment to see if the human
parkin protein could rescue the Drosophila parkin-
phenotype. Perhaps in the hope of examining human AR-JP more closely they could
attempt to find the Parkin gene ortholog in mice, using the same method to find
the gene: screening the genome for sequence and structural similarities.
Experiments should also be designed to see which proteins interact
with parkin. This can be done by immunoprecipitation, then any proteins
found should also be tested for their effects on the parkin- phenotype.
It could be that the parkin loss causes another protein to lose its
function, and this is the protein that is key to mitochondrial function. It
is also important to discover the main function of the parkin protein
so that we can understand the mechanism by which the mitochondrial pathology
and cell death are triggered and carried through. The Parkin gene could
also be cut at varying intervals to determine which parts of the sequence are
necessary for proper functioning of the protein and normal phenotypes: perhaps
it is only the amino-terminus, or the ring-finer domain that interact with the
necessary proteins or targets.
Another experiment could be to over express the parkin
protein and examine the phenotype that results from this, perhaps over expression
of the protein could cause other dysfunctions that might be helpful in determining
the role of parkin during normal functioning.
Comments, questions, and suggestions to: Megan McDonald