On 2019-11-05 16:43:34, user Cyrus Cheung wrote:
BI598 Group 6 (Alondra, Cecar, Cyrus, Safiya, and Can)
A review written by Boston University undergraduate students majoring in Neuroscience and Neurobiology as a requirement for the class, Neural Circuits (BI598).
Summary:<br />
Multiple Sclerosis (MS) is an inflammatory, neurodegenerative disease of the CNS characterized by both grey and white matter injury. Hammond et al used the established experimental autoimmune encephalomyelitis (EAE) animal model to recapitulate features of the disease. Their goal was to evaluate whether complement dependent synapse loss contributes to grey matter degeneration in EAE, which they highlight as being understudied. The complement system is part of the innate immune response and it signals microglia to prune synapses by phagocytosis. Therefore, their experiments revolved around studying the effects of knocking out C1qa and C3, which are both immune signaling proteins that signal microglia to phagocytose cells and cause neurodegeneration.
To determine if EAE hippocampus produces the complement proteins that could make synapses vulnerable to phagocytosis by glia, they analyzed C1q and C3 protein and mRNA expression by Western blot and qPCR. As shown in Figure 1A-C, they determined that the increase in protein expression of C1q and C3 in the EAE hippocampus was due to local gene expression, and not because of blood brain barrier breakdown. They also isolated CD11b+ microglia from the hippocampus and cortex, and found that CD11b+ cells in EAE mice overexpressed C3 28 days post immunization. However, they found no significant difference in the C1qa expression (Figure 1D). They concluded that microglia contribute to elevated C3 expression, and possibly C1q expression. They acknowledge, however, that the EAE elevated expression of complement proteins could be induced by other cells in the hippocampus.
They performed IHC using anti-C1q and anti-C3d antibodies to investigate the locations of elevated C1q and C3. Through IHC, they measured an increase in C1q fluorescence across the hippocampus in EAE vs sham mice (Figure 2B). At high (60-100x) magnification they observed that C1q was diffusely localized throughout the neuropil but also was localized at higher density in small punctate regions, some of which co-localized with synapses or along the dendrites in both sham and EAE brains (Figure 2D-E).
To answer whether the loss of C1q or C3 protects against EAE induced motor impairment caused by spinal cord damage, the authors compared the clinical score of WT EAE mice with the C1qa and C3 KO from 0 to 26 days post induction (Figure 3). The graph shows that only the C3 KO presented less severe motor deficits, C1qa KO did not alter the course of the EAE disease, and neither the C1qa KO nor the C3 KO changed the symptom onset.
The researchers proceeded to study how knocking out C1qa or C3 in the CA1-stratum radiatum layer of the hippocampus could have a protective role for synapse loss. They used IHC to stain for two postsynaptic markers: Homer1 and PSD95 in mice 28-30 days post immunization. As seen in Figure 4C-D, the knockouts played a slight protective role in synapse loss, with C3 KO playing a greater effect.
Hammond et al. took their results further by investigating whether C1qa or C3 KO alter the morphometric parameters of microglial activation induced by EAE. More specifically, they utilized IHC to stain for microglial proteins IBA1 in the CA1-SR of the hippocampus (Figure 5). They then tested for various parameters such as sum volume, sum intensity, surface area, and skeletal length as a way to test for change in cell morphology. The data from this figure suggests that the C3 KO prevented hippocampal microglial activation induced by EAE, however it also revealed that the C1qa KO had no effect on microglial morphology.
Merits: <br />
The paper identifies an interesting gap in knowledge, the introduction highlights their goal of studying the role of grey matter degeneration, a result of complement-dependent synapse loss. The complement-dependent synapse loss was studied effectively by use of C1q and C3 knockout mice.
Overall, they made a convincing argument that the C3 knockout has the most potential for any rescue of symptoms. This makes sense because C1q activates C3, which accumulates at synapses, hence microglia can detect it as a signal to phagocytize. It is intuitive to think that knocking out C3 would slow the engulfment by microglia and decrease the loss of synapses in EAE.
Specific Critiques:<br />
Overall, the paper made a connection between multiple sclerosis and the complement pathway with a focus on the motor cortex. However, multiple sclerosis has many different symptoms beyond just motor movements, therefore the paper would benefit from an investigation into other symptoms and circuits of MS, such as vision and sensation. In the introduction, the paper also discusses the use of the induced EAE mouse model as a model of MS grey matter injury; however, the results section does not indicate the data refers to grey matter. An explanation of how the complement dependent synapse loss affects grey matter would tie the paper together.
They reported that the inflammatory response was localized in the hippocampus through the qPCR analyses that showed local gene expression of C1q and C3. Overall in Figure 1 the error bars were large, the data would specifically benefit from a larger and more consistent sample size, since the N’s range from n= 5 to n=11. While Figure 1B-C quantifies levels of C1q and C3 expression in the hippocampus, Figure 1D shows data from CD11b+ cells isolated from the cortex and hippocampus, which was as stated, a way to “obtain sufficient cells”. In order to discern if the increased expression of C1qa occurred locally in the hippocampus or in both hippocampus and cortex, the author should include another set of graphs for the qPCR results of C1qa and C3 gene expression of just the hippocampal extracts separated from the cortex. Without this, it is inconclusive if microglia contribute to elevated C3 expression in EAE, as there is insufficient evidence of microglia directly elevating the levels of C1q and C3 in the hippocampus.
The method of evaluating the protein levels of C1q and C3d in the hippocampus of EAE mice was well thought out. However, in order to make a stronger claim a few elements could be added to Figure 2. In addition to the treated hippocampal areas, a normalized signal should have been presented such as a DAPI stain in order to show the living cells in the localized area, allowing for a helpful comparison. An alternate would be to stain for blood vessel markers, allowing for the reader to understand the layout of the image presented. On the other hand, a simple addition of insets on top of the current images would give the reader a greater sense of what exactly they are looking at. This would allow for quantification of the data presented as the colocalization is not very clear. The C3 knockout would also benefit from an increased sample size, since this might make the large error bars smaller. A discussion of a discrepancy between a significant increase of C1q, but not C3, in the EAE mice and it’s relation to its synapse loss significance that is discussed in later figures would help bridge the paper together and explain something that might be unexpected.
Figure 3 opens the possibility of targeting C1q and C3 to ameliorate the motor deficits caused by MS through investigation of the different complement pathways. However, the paper did not elaborate on why they concluded the alternative complement pathway as the most important in EAE. As stated in the introduction the complement system, classical, alternative, and lectin pathways all converge on the production of C3. Therefore, there is not enough evidence to conclude that the lectin and classical pathways are also not involved. An explanation on how they conceptualize these pathways would provide increased clarity to the conclusions made from the data.
Through the pictures and graphs shown in Figure 4, the authors are trying to show that knocking out C1qa and C3 decreases the synapse loss seen in EAE. The data would benefit from an increased sample size, especially for the C3 KO model, as there are large error bars. They use Homer1 and PSD95 as postsynaptic markers but they should confirm the presence of synapses by including EM images or using a presynaptic marker such as Synapsin, in addition to Homer 1/PSD95. Authors could also run a western blot for synaptic proteins like DLG4, Synaptophysin or Neuroligin to confirm the presence or absence of synapse through the different conditions. Again, the data does not provide enough evidence to conclude that the alternative pathway is the most important in EAE. A more elaborate explanation of how the data concludes it is the alternative pathway would add clarity.
Figure 5 displays the stained brain sections from the WT, C1qa KO, and C3 KO, sham and EAE immunized mice along with the quantification of various parameters to show that C3 KO mice with EAE have reduced microglial activation when compared to the WT EAE mice. To properly convey this, a few sentences on how C3 is specifically connected or related to microglia are needed. Considering the IBA1 stains can be expressed in other parts of the brain, it would be beneficial to check for other markers such as TMEM 119, a microglial cell surface protein that is not expressed in other neuronal macrophages or immune cell types. In addition to this, figure 5A, could potentially be combined with figure 1 considering they both look at morphology.
Minor Concerns: <br />
Introduction<br />
The introduction includes a few grammatical errors:<br />
In the second paragraph, “triggers” should be “to trigger”<br />
In the fifth paragraph, the authors should have used present tense while writing about the previous findings.
Methods<br />
They included background information about C1qa in the methods section. This belongs to the introduction section of the paper.
The clinical score was mentioned in the methods and it would be beneficial to name the section appropriately or create an entirely new section.
It was unclear why Homer1 and PSD95 was chosen to investigate synaptic loss in the CA1-stratum radiatum. Meanwhile, IHC with CamKII is more commonly used and could show synapses more clearly. More explanation could be provided.
In the fourth line of the EAE section, instead of “each of two sites”, it should be “each of the two sites”.
Figures<br />
In Figure 1D, the authors include data on the qPCR results of C1qa and C3 from CD11b+ microglia/myeloid cells and mention this in the results. However, they should clarify the role and importance of CD11b+ microglia/myeloid cells by adding background information in the introduction.
Figure 2 and Figure 4 could be visualized better with a larger and higher resolution image.<br />
Figures 4 A through B are stained in different colors and the green is harder to see than the grey. It is recommended that they stick to one color so that they both images are easy to see.
Results<br />
In the fifth line, “C1q and C3 protein” should be “C1q and C3 proteins”<br />
In the sixth line, the sentence should start as “By Western Blot, we found that...”<br />
Please review the paper for spelling, grammatical, and punctuation errors.
Future Directions:<br />
The researchers investigated a motor impairment in a mouse model with C1q and C3 knockout, affecting the hippocampus as well. Different diseases, such as Alzheimer’s Disease, are heavily associated with hippocampal damage, indicating the complement pathway could be extrapolated as a therapeutic target for neurodegenerative diseases as a whole. Behavioral paradigms that test other symptoms of multiple sclerosis, such as vision, and tasks that test for Alzheimer’s Disease, such as working memory, could show the impact of the complement pathway on common diseases.
As complement genes are important for neuronal development, confounding factors might have played a role in the experimental paradigm. Use of ASO to knock down genes and bypass development would help remove confounding variables in the experiment.
Lastly, accounting for more time points prior to 28-30 days post immunization would allow for a clear visualization of the EAE progression. Including more time points for figures 4 and 5 would allow for a comparison of how synapse loss progresses through time.