Increased caspase-9 staining was present in Ster retina only, with a peri-nuclear staining predominantly in the lower part of the ONL. To match the various RD models and their very different degeneration kinetics with each other, all values were expressed as logarithm to base
These RD models harbour genetic defects mostly affecting the phototransduction cascade but include also such which are disturbing the visual cycle Rpe65 KO and the structural integrity of the outer segment rd2. As a result, the comparative analysis of characteristic cell death processes for the first time highlights the over-riding importance of a common, alternative mechanism for photoreceptor degeneration.
To put our report in a perspective, many studies on cell death in the retina and other parts of the central nervous system have previously resorted to tissue based methods e. Such methods are particularly useful in conditions where there is a homogenous cell population and a highly synchronized onset of cell death and are thus ideal, for instance, for cell culture. For our analysis, we thus focussed on methods that afforded cellular resolution to be able to unequivocally attribute cell death related processes to primary photoreceptor death and to distinguish these processes from secondary or tertiary events.
Previous studies on cell death in hereditary retinal degeneration have often suggested apoptosis as the main degenerative mechanism  ,  , . These earlier studies, however, based their conclusion on analysis methods now known not to discriminate between apoptosis and other forms of cell death. For instance, the TUNEL assay, originally thought to be a marker for apoptosis  , generally labels all kinds of dying cells, including necrotic cells .
Apoptosis may be defined as an active process resulting in orderly self-disintegration of a cell. Hallmark features of apoptosis include an up-regulation of pro-apoptotic genes and proteins, such as the transcription factor c-fos and in particular Bcl-2 family proteins such as BAX, which participate in forming the mitochondrial permeability transition pore MPTP , allowing mitochondrial proteins including cytochrome c to enter the cytoplasm.
Cytoplasmic cytochrome c aggregates with apoptotic protease-activating factor APAF and caspase-9 to form a multimeric protein complex termed the apoptosome . This complex then cleaves and activates down-stream executioner caspases such as caspase Classical apoptosis occurs during retinal development until about 3—4 weeks post-natal .
Indeed, developmental apoptosis temporally coincides, at least partially, with mutation-induced cell death . This introduces a confounding factor which may explain some of the contradictory reports in the literature. Our study demonstrates that wild-type photoreceptors are capable of executing apoptosis at least until P42; by contrast, however, we see that mutant photoreceptors normally take a non-apoptotic route as a means for orderly self-destruction.
Importantly, therapeutic strategies based on the inhibition of the apoptotic cascade have had little success or produced conflicting findings. For instance, neither the pharmacological inhibition of the caspase cascade  , nor the genetic manipulation of Bcl-2 and Bcl-XL  , c-fos  , or caspase-3  promoted long-term photoreceptor survival. At present it is not clear whether these findings relate in part to developmental cell death see above or would have been interpreted differently if the study  had also included observations of a model with a much stronger BAX response, such as the Ster rat investigated by us.
At any rate, our results do not show any evidence for major BAX activation in degenerating retina, with the notable exception of Ster photoreceptors.
In recent years a growing body of evidence has suggested the activity of alternative cell death mechanisms in RD  ,  — . The analysis of such mechanisms faces the major obstacle of identifying alternative and causative metabolic processes. Excessive cGMP signalling was associated with a strong increase in enzymatic activities of calpain-type proteases  , PARP  , and HDAC  , which we found to be causally involved in photoreceptor cell death.
Calpain activation, which was also seen by others in different RD models  , is a well-established phenomenon in necrosis and alternative cell death mechanisms  , . While HDAC and PARP enzymes are ubiquitously expressed and involved in epigenetic gene regulation and DNA repair  , respectively, their excessive activation has repeatedly been connected to alternative mechanisms of neuronal cell death  — .
We found that all these processes were also involved in RD caused by the different mutations, in various genes and in both mouse and rat. Importantly, the cellular resolution afforded by the used assays allowed clear distinction between cells dying an apoptotic death and cells dying through an alternative pathway. This could suggest that the latter two relate to early metabolic processes in the execution of cell death.
Together with other earlier data  ,  ,  ,  our present findings prompt us to propose a potential pathway for cGMP-induced cell death: Both routes Figure 6 act in unison to drive a photoreceptor cell to its demise, but, surprisingly, this alternative form of cGMP-induced cell death appears to be 4—6 times slower than apoptosis .
Importantly, the presence of this pathway and the connections between the different metabolic processes were confirmed by interventional experiments in the rd1 mouse demonstrating the neuroprotective effects of inhibition of PKG  , calpain  , PARP  , and HDAC .
Classical apoptosis, such as it occurs in Ster transgenic photoreceptors, involves a mutation-induced up-regulation and translocation of BAX protein to form the mitochondrial permeability transition pore MPTP. This leads to leakage of cytochrome c from the mitochondria to the cytoplasm, where it combines with apoptotic protease activating factor APAF and caspase-9 to form the apoptosome, which in turn activates down-stream executioner caspases, including caspase Importantly, this alternative, non-apoptotic cell death mechanism offers a number of novel targets for neuroprotection of photoreceptors.
The observed PARP activity deserves some additional considerations: In classical apoptosis the PARP enzyme is cleaved and inactivated by caspases, resulting in a specific 85 kDa PARP fragment, the presence of which is often used to characterize apoptosis as such . Hence, what we found in mutant photoreceptors is the exact opposite of what would happen in apoptosis, which thus provides further evidence for a non-apoptotic photoreceptor cell death, an alternative cell death mechanism that could share some features with PARthanatos .
The fact that photoreceptors use a non-apoptotic mechanism when in principle they are capable of executing apoptosis raises the question as to what the physiological and even evolutionary advantage of this mechanism may be. Apoptosis is a process that requires energy in the form of ATP . The insult caused by a genetic mutation may exhaust such energy resources to the point that apoptosis can no longer be executed. Necrosis on the other hand would result in inflammation and could cause additional extensive tissue damage.
Hence, it may make sense for a cell to execute the slow, alternative and probably ATP-independent pathway laid out here to limit the damage to the surrounding neuronal tissue. An important consequence of the high genetic heterogeneity of retinal degenerations is that for any pathogenic mutation there may be only a very low number of patients  , . This calls for the development of mutation-independent treatments that could address larger groups of RD patients.
In the context of rare retinal diseases, such treatments appropriate for a large number of patients may dramatically improve the perspectives for both a successful clinical translation and the commercial viability of corresponding drugs. We found that the alternative cell death mechanism described above was active in all investigated animal models. Of particular importance for this mechanism may be the observed accumulation of cGMP in mutant photoreceptors. While this was already known for retina suffering from mutations in Pde6b and Pde6c i.
A potential explanation for this remarkable phenomenon in rhodopsin mutants could be either the longer life-times of activated rhodopsin resulting in a stimulation of cGMP synthesis and an increase in net cGMP  or a failure to activate downstream PDE6 in cases where rhodopsin is absent i.
While these findings highlight cGMP-signalling for the development of novel neuroprotective treatments, there is one exception: Indeed, here, unliganded opsin was proposed to cause a constitutive activation of phototransduction and hence low cGMP-levels . On the other hand, since all further down-stream processes appear to be the same in all mutants investigated, a disruption of the visual cycle by Rpe65 KO  might cause minor elevations of cGMP — perhaps below the detection levels of our immunohistological methods — and still trigger cell death.
Mutations in the same gene may potentially trigger distinct degenerative processes . Our study more extensively shows how intragenic variability of RD mutations may initiate different cell death mechanisms: The recessive rd1 and rd10 mutations in the Pde6b gene result in activation of the same non-apoptotic pathways.
While all three mutations reside in the rhodopsin gene, the concurrent activation of apoptotic and non-apoptotic cell death observed in the Ster situation suggests that human patients with similar mutations may need combination therapy targeting both degenerative pathways simultaneously. Likewise, since we found that photoreceptors wild-type are in principle able to execute apoptosis, we cannot exclude the possibility that under circumstances in which non-apoptotic cell death is blocked, the cell may switch to apoptosis.
This possibility needs further investigation and might also require the development of combination therapies. Another question, that will be important to address in the future, relates to the fact that all mutant photoreceptors carry a genetic defect that will eventually destroy them.
Yet, the time-point at which a mutant photoreceptor dies appears to be entirely random, and, in the human situation, the time from the first to the last photoreceptors' death may cover many decades .
The exact reasons for this phenomenon are unknown but could be explained by stochastic effects similar to what is seen in the decay of radioactive elements . This opens the possibility that even a minor shift in the dynamics of these stochastic processes — such as interference with processes like those studied here — could improve photoreceptor survival dramatically. In conclusion, this work demonstrates the existence of a common, non-apoptotic cell death mechanism for hereditary photoreceptor degeneration.
The tentative cell death pathway laid out here Figure 6 provides a number of novel targets for neuroprotective treatment approaches  ,  ,  ,  ,  and, importantly, a unifying principle for RD caused by a variety of different mutations in different genes.
As such, this common cell death pathway may be of major importance for future RD therapy developments and possibly for also other neurodegenerative diseases. Correlation of selected cell death markers to loss of photoreceptors, related to Figure 1.
In all three models, calpain activation peaked together with the TUNEL assay, and correlated with the strongest loss in the number of photoreceptor rows. The grey area indicates the loss of photoreceptors. Throughout the retinal degeneration, activation of caspase-3 was absent in rd1 and P23H retina, but present in Ster retina. Values are mean from at least three different animals.
Expression of activated BAX in wild-type, rd1 and Ster retina. In wild-type mouse retina at P11 left panel , a mouse monoclonal antibody directed against activated BAX clone 6A7 detected positive cells only rarely, but then in all layers of the retina.
In rd1 mouse retina at P11 — the onset of RD in this model — activated BAX is detected only very rarely, with BAX detection levels very similar to age-matched wild-type middle panel; cf.
In contrast to this, in the outer nuclear layer ONL of P12 Ster rat retina, the BAX antibody immunodecorates mitochondria, in particular in individual photoreceptor inner segments, synaptic terminals, and perinuclear areas right panel. This mitochondria specific staining pattern in Ster retina is consistent with the reported role of BAX in the formation of the mitochondrial permeability transition pore and the initiation of apoptosis.
Images are representative for immunostainings obtained from at least three different animals for each genotype. Note that use of secondary anti-mouse antibodies led to an unspecific IGG decoration in inner retinal blood vessels in mouse tissues see asterisks in wild-type, rd1. Cell death markers in wild-type mouse retina.
Well-type retina occasionally showed cells positive for both apoptotic and non-apoptotic cell death markers A. As the number of positive cells is rather small, please note that the pictures shown are selected not as the representative but somewhat an exaggeration of the real number of dying cells.
Heat map representing metabolic activities in corresponding wild-types B , similarly as in Figure 5 for RD mutants, shows that cell death during wild-type retina development displayed activation of both apoptotic and non-apoptotic pathways. See also Table S2. Numbers given represent mean values for the percentages of positive cells for each marker, followed by standard error of the mean SEM , and p-values for comparisons with corresponding, age-matched WT.
Significant differences between RD mutants and WT were found almost only for non-apoptotic processes, with the notable exception of the Ster mutant where also apoptotic processes were significantly activated. Note that in contrast to Fig. For each genotype, at the respective peak of degeneration, the percentage of cells positively labelled for the various cell death processes is given as mean value, followed by SEM, and number n of different specimens analysed.
To assess the relative importance of these processes for retinal degeneration the percentage of TUNEL positive cells is also given. LaVail for transgenic rats; T. Haq for helpful comments and discussions; S. Abdshill for skilful technical assistance. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
National Center for Biotechnology Information , U. Published online Nov Author information Article notes Copyright and License information Disclaimer. The authors declare that they have no conflict of interest. Received Jun 12; Accepted Oct This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
This article has been cited by other articles in PMC. Abstract Cell death in neurodegenerative diseases is often thought to be governed by apoptosis; however, an increasing body of evidence suggests the involvement of alternative cell death mechanisms in neuronal degeneration. Introduction Apoptosis is a programmed cell death mechanism that is often invoked for neurodegenerative diseases.
Open in a separate window. RD animal models used and their genetic defects. Materials and Methods Animals Animals were housed under standard white cyclic lighting, had free access to food and water, and were used irrespective of gender.
Table 1 List of animals used, genes affected, and original references where applicable. CD SD wild-type - - Crl: Italic fonts indicate mutant name or affected gene. Table 2 List of antibodies used in this study. Calpain in situ activity assay Calpain activity was investigated with an enzymatic in situ assay . Microscopy, cell counting, and statistical analysis Light and fluorescence microscopy were usually performed at room temperature on an Axio Imager Z. Results In RD models the peak of cell death varied depending on severity of genetic insult To study the cell death mechanisms governing RD, we first performed a detailed analysis of the temporal progression of the degeneration for each of the 10 animal models used Figure 1.
Progression of cell death in inherited RD models. Apoptosis was restricted to degenerating Ster retina We looked for increased expression, localization, or activation of Bcl-2—associated X protein BAX , cytochrome c, cleaved, activated caspase-9 and -3 Figure 3 , quantification in Table S1 and S2. Apoptosis in the retina is restricted to the Ster rat model. Non-apoptotic cell death in photoreceptor degeneration We have previously shown that rod photoreceptor degeneration in rd1 mice is characterized by accumulation of cyclic guanosine monophosphate cGMP , increased activities of histone deacetylases HDAC , poly-ADP-ribose polymerases PARP , and calpains  ,  , .
Cell death in hereditary retinal degeneration is predominantly non-apoptotic. Heat map representing metabolic activities in different RD models. Discussion Our study provides a detailed and comprehensive overview of the temporal progression and the kinetics of cell death in ten different, commonly used RD animal models. Apoptosis during retinal degeneration Previous studies on cell death in hereditary retinal degeneration have often suggested apoptosis as the main degenerative mechanism  ,  , .
Alternative cell death mechanisms In recent years a growing body of evidence has suggested the activity of alternative cell death mechanisms in RD  ,  — . Two routes to cell death. Perspectives for mutation-independent RD treatment An important consequence of the high genetic heterogeneity of retinal degenerations is that for any pathogenic mutation there may be only a very low number of patients  , .
Supporting Information Figure S1 Correlation of selected cell death markers to loss of photoreceptors, related to Figure 1. TIF Click here for additional data file. Figure S3 Cell death markers in wild-type mouse retina. Acknowledgments We thank M. In URT-based pulse-chase assays, the labeled test protein is quantified by measuring its level relative to the level of a stable reference at the same time point.
The logic of these assays Piatkov et al. In addition to being more accurate than pulse-chases without a stable reference, URT assays make it possible to detect and measure the degradation of a test protein during the pulse before the chase Piatkov et al.
The magnitudes of the initial pre-chase degradation of these fragments were similar to that of Arg -TDP43 f. For designations, see the legend to Figure 1B. The arrowhead indicates the site of cleavage by deubiquitylases Figure S1C.
The mCherry moiety of mCherry-Ub R48 was detected by red fluorescence, and the Asp -TDP43 f fragment was detected by indirect immunofluorescence, using anti-flag antibody and a fluorescein-conjugated secondary antibody.
The error bars indicate SEM standard error of measurement. Immunofluorescence microscopy indicated an increased in vivo aggregation propensity of TDP43 fragments compared with full-length TDP43 Furukawa et al. The use of URT made it possible to unambiguously identify transfected EF cells through their red fluorescence, irrespective of whether or not these cells were capable of Nt-arginylation, i. Arrowhead indicates the metalloprotease cleavage site.
D Domain organization of human Tau-2N. Arrowheads indicate the calpain cleavage sites. A Human APP near its transmembrane domain, indicated by yellow rectangle. Lane 2, same as lane 3 but without Ate1. In agreement with S. Lanes 1 and 2, FA extracts from brains of non-Tg and 5xFAD mice, respectively, that were treated with inhibitors of proteasome and neprilysin 12 hrs before harvesting brains. Lane 5, same as lane 2 but from 5xFAD mice that had not been treated with protease inhibitors.
In contrast to human brains, for which this interpretation remains a conjecture, it was strongly supported by assays with brains from 5xFAD mice Figure 5E , lanes 2 vs. Tau and its fragments are associated with neurodegenerative disorders that are referred to as tauopathies and include AD and many cases of FTLD Prusiner, ; Selkoe, Tau can be cleaved by calpains or caspases.
The resulting fragments tend to be more aggregation-prone and toxic than full-length Tau Spires-Jones et al. Through a decrease in the in vivo levels of an aggregation-prone protein, its selective degradation would slow down the formation of an aggregation-nucleating protein oligomer, a key intermediate in the growth of larger aggregates. Together with our findings about natural fragments of other proteins Piatkov et al.
A plausible explanation is that the initial oligomerization of a protease-generated C-terminal protein fragment e. This oligomerization may occur rapidly enough after the protease-mediated cleavage of a full-length precursor e.
The rest of this fragment, i. The latter process can account for negligible levels of, e. An alternative possibility is that the cleavages giving rise to Nt-arginylatable fragments may take place after the formation of aggregates, i.
Possible fates of N-terminal protein fragments are considered below. If so, it is possible that the number and initial levels of protease-generated protein fragments in a cell are significantly larger than is currently observed, i.
Its down-regulation during aging Gabius et al. These recently discovered degrons, created by Nt-acetylation Hwang et al. Its many functions include protein homeostasis, in conjunction with the N-end rule pathway. Plasmids, constructed by standard methods, are described in Tables S1 and S2. Care and treatments of mice were performed according to the relevant NIH guidelines, as described in Supplemental Experimental Procedures. Udartseva for genotyping mouse strains and C. Rosen for help with Ate1 isoforms.
We are also grateful to members of the Varshavsky laboratory for their assistance, and to S. Gutierrez for their help, advice and support at the mouse transgenic facility. This study was supported by grants to A. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Supplemental Information includes Supplemental Experimental Procedures, four figures and two tables.
National Center for Biotechnology Information , U. Author manuscript; available in PMC Apr Brower , Konstantin I. Author information Copyright and License information Disclaimer. The publisher's final edited version of this article is available at Mol Cell. See other articles in PMC that cite the published article. Associated Data Supplementary Materials Open in a separate window.
E Quantification of D. G Quantification of F. See also Figure S1. B Quantification of A. See also Figure S2A. F Quantification of E.