Advances in the identification of the molecular mechanisms behind different autophagic pathways during recent years and the increasing number of connections between autophagic dysfunction and disease have generated growing interest in the study of autophagy. One of the major reasons why the study of CMA has been limited for years to only a handful of groups has been due to the fact that this autophagic pathway could only be monitored using specialized techniques that are not routine in most laboratories. Until now, the best method to determine changes in CMA activity has been by measuring the ability of isolated lysosomes to take up well-characterized CMA substrates 16
. CMA activity has only been indirectly inferred in cells through the combination of different approaches, which include, among others, metabolic radiolabeling and measurement of protein degradation in the presence of different protease inhibitors, and tracking of the number and intracellular localization of the subset of lysosomes active for CMA 16
. The genetically-encoded photoconvertible PS-CFP2-based reporter described in this work should make monitoring of CMA in cells in culture broadly accessible and allow tracking the activity of this autophagic pathway in primary non-dividing cells, from which isolation of lysosomes is not feasible.
Accumulation of tagged constitutive fluorescent proteins upon blockage of proteolysis has been extensively used to investigate the activity of the ubiquitin/proteasome system and macroautophagy. Using photoconvertible proteins allows similar analysis without having to block protein synthesis and degradation 18,19
. An added advantage of photoconversion in the case of the CMA reporter is that it permits detection of the protein associated to the cytosolic side of the lysosomal membrane (green) against the cytosolic pool that, for the most part, is in a different fluorescence state (cyan). Furthermore, the use in our study of the new generation of fluorescent proteins (CFP, mCherry) that preserve their monomeric characteristics inside the cells 17
has helped in eliminating problems that could result from oligomerization of fluorescent proteins. This was particularly important because, it has been previously reported that protein oligomers formed by GFP-like proteins, such as DsRed and mRFP1, can be delivered to lysosomes, likely by macroautophagy- or microautophagy-related mechanisms 21
. We are confident that macroautophagy does not contribute significantly to the lysosomal association of our fluorescent reporter because as shown in and , blockage of macroautophagy did not have any significant effect on the amount of KFERQ-PS-CFP2 associated with lysosomes. Regarding microautophagy, we have recently reported that in mammals this process occurs preferentially in endosomes and it is independent of the presence of LAMP-2A 26
. Because we did not observe colocalization of KFERQ-PS-CFP2 with endosomes () and knock-down of LAMP-2A effectively eliminated its association with lysosomes () makes contribution of microautophagy unlikely and supports that this novel reporter is delivered as monomers to lysosomes by CMA.
KFERQ-PS-CFP2 does not co-localize completely with classic lysosomal markers such as LAMP-1 and LAMP-2, this is likely because these LAMPs are also present in endosomes (where CMA does not occur), and that not all cellular lysosomes are capable to perform CMA (although they all contain LAMPs), The unusually low level of colocalization of KFERQ-PS-CFP2 with LAMP-2A-positive structures could be explained by the same reasons as above, but in addition to the antibody used to differentiate the LAMP-2A variant from the other LAMP-2 proteins binds the same region as the CMA substrates (the 12 amino acids of the cytosolic tail). Therefore, staining for LAMP-2A will be lower when KFERQ-PS-CFP2 is bound to its cytosolic tail as a result of activation of CMA.
Previous analysis of KFERQ-containing proteins in different organs by immunoblot have revealed organ-specific differences in the abundance of these proteins during starvation, leading to proposed possible cell type-dependent differences in CMA activity 27
. However, direct comparison of CMA activity across different cell types has not been possible until now. Our studies with the photoconvertible CMA reporters reveal that, indeed, the ability of cells to upregulate CMA in response to serum removal varies depending on the cell type. Furthermore, we have found that their basal CMA levels are also different. Future follow-up of these findings may help elucidate if differences in the normal contribution of CMA to protein turnover among neuronal types is the reason behind their different susceptibility to blockage of this pathway by pathogenic proteins as is the case, for example in Parkinson’s disease.
The coexistence of different proteolytic pathways inside cells and the fact that some proteins can undergo degradation by several of them supports that although not redundant, these pathways can compensate for each other. Indeed, previous studies have revealed the existence of a cross-talk between the two stress-induced forms of autophagy – macroautophagy and CMA – and between macroautophagy and the ubiquitin/proteasome system. Through the use of the photoconvertible reporters, we have now identified a new point of interaction between CMA and the proteasome. The ability to monitor CMA activity in intact cells makes it now possible to start addressing if this cross-talk is universal, the duration of the compensatory activations and whether the ability to upregulate CMA or not in pathological conditions with compromised proteasome activity determines the rate of disease progression.
Overall, as for any other image-based procedures, the use of the photoconvertible CMA reporters would be maximally strengthened by performing kinetic, rather than single-point analysis, and applying rigorous morphometric procedures. In addition, when used for the first time in a new cell type or under specific conditions, it is highly advisable to include the following controls in a pilot assay before the experimental analysis: checking for autofluoescence in untransfected cells; determining the kinetics of expression of the reporter (in the case of studies with transiently transfected cells); analyzing the efficiency of photoconversion under the standard conditions described here and optimizing these parameters for the specific cell type if needed; and analyzing of cell viability after photoconversion (to determine possible toxic effect of the 405nm light on specific cell types). We also recommend the following approach to validate that the formation of fluorescent puncta with the CMA reporter results indeed from the delivery of the fluorescent protein to lysosomes by this pathway: confirming that the puncta correspond to lysosomal location (by colocalizing with LysoTracker or lysosomal components such as LAMPs); discarding contribution of macroautophagy to the lysosomal delivery (by demonstrating absence of colocalization with LC3 or lack of effect of treatment with 3-methyl adenine in the final number of puncta); and confirming the dependence of the punctuate pattern on CMA (by demonstrating absence of fluorescent puncta in LAMP-2A-deficient cells). In summary, we have developed a photoconvertible fluorescent reporter that allows monitoring and quantifying CMA activity in both transformed and primary cells in culture and have validated its applicability for high-content microscopy.