Joanne (5)

My quiz was on the Hillman et al. paper. I attached my quiz results above. It was an open book quiz asking us to interpret data and give our own opinion about the data and the author’s conclusions. They studied the expression of UBE3A in the CNS and non CNS to see if there is genomic imprinting that regulates the dosage of UBE3A in neurons. I chose this as my top 4 assignment because it was my first time coming up with my own opinion that would go against the authors’ conclusion. The authors had the assumption that imprinting results in a reduced dosage which lead to the conclusion that imprinting doesn’t regulate dosage of UBE3A. However, UBE3A is clearly higher in the CNS than non CNS, so if we think of this case in another way, apart from their assumption, genomic imprinting may be associated with higher level of UBE3A in the CNS.

Below is the worksheet that my friends and I worked on in preparation for the quiz. [there was misformatting when transferring the document such that the numbers are off]

Figure 1: expression of UBE3A in mouse and human CNS (eutherian)

1. What is FPKM normalization?
   • This stands for fragments per kilobase million from the RNA seq data and it’s a a way of reporting RNA seq results. Normalized here just means that it is per million bases so is is a normalized number rather than being per some other number that is different between RNA seq data (want all data to be reported using same units)
   • Why is there no significance of ube3a in hippocampus in FPKM normalization?
     • There is a significant difference between all samples but just not specifically between hippocampus compared to liver and thymus (but there is a sig diff between hippocampus and heart and lung)
2. What question is addressed in the experiment?

To test if imprinting in neurons (in the paternal allele) of UBE3A reduces dosage of UBE3A in the neurons compared to other cell types  

1. Why did the authors have a Ube3am+/p-?
   • To estimate the relative expression levels of the maternal Ube3a allele (to control for the potential effects of the paternal allele on the expression levels)
2. Justify their choices in technique, control.. Etc.
3. What could the authors have done for better results?
4. What do the data show?

• Higher Ube3a transcript level in the CNS compared to the N-CNS in both Ube3a m+/p+ and Ube3aM+/p-
• a) Ube3a has significantly higher transcript levels in the CNS (cortex, hippocampus) compared to other cells (heart, kidney, etc) in both Ube3a m+/p+ and Ube3aM+/p-
• b) Ube3a has significantly higher protein levels in the CNS (cortex, hippocampus) compared to other cells (heart, kidney, etc) in both Ube3a m+/p+ and Ube3aM+/p-
• c) There is no significant difference between the CNS of Ube3a m+/p+ and Ube3aM+/p- in both the protein and transcript expression

1. What can we directly conclude from the data?
   • Functional WT paternal Ube3a allele is not necessary to have WT levels of Ube3a transcript in both the CNS and N-CNS tissues (fig 1a)
   • Though hippocampus tissues have a large 5 fold difference between maternal allele and paternal allele transcript abundance, its overall Ube3a transcript count is comparable to other tissues, like the thymus, heart, and liver.
2. In one sentence, what did the authors demonstrate?
   • CNS transcript abundance of Ube3a in human and mice are similar to many Non-CNS tissues and the maternal Ube3a allele produces the bulk of the transcript product in  CNS.
3. Are these results what you would expect? Why?
   • In some ways yes and in some no. I would expect to see equal or lower expression levels of Ube3a in CNS compared to N-CNS since in the CNS, Ube3a is imprinted and therefore there is only expression coming from the maternal allele, however, we observe that Ube3a is expressed in higher amounts in the CNS compared to N-CNS. This suggests that Ube3a is being overexpressed in the maternal allele or there is some sort of regulatory mechanism happening when the paternal allele is expressed. Other results are expected.

 

Figure 2: Maternal Ube3a compensates for loss of paternal Ube3a expression during neurogenesis

1. What experiments did they do? What are the controls and why did they have that?

a)Immunofluorescence imaging of prenatal primary neurospheres

Maybe no control:  they compared the pattern in paternal and maternal Ube3a

1. b) western blot to show the difference in YFP and normal protein
2. d) measured intensity values  

1. What was the purpose of these experiments? What question were they addressing?

• This was to show that imprint is acquired during neuronal differentiation and there is bias toward maternal allelic expression of Ube3a in the days post differentiation
• Question addressed: does the maternal allele shift expression in a compensatory way for the paternal decreased expression?

1. What do the data show?

16. d) intensity of maternal UBe3a increases significantly at 16DPD. Paternal decreases in a linear manner from DPD1 to DPD16. There seems to be an inverse relationship between the maternal and paternal Ube3a in the primary neurons
17. e) At DPD1, the difference in intensity of Ube3a is not significant. As DPD increases, the discrepancy between maternal and paternal protein intensity increases.

1. What can we conclude from the data?

There seems to be an inverse relationship between the maternal and paternal Ube3a in the primary neurons. Intensity of maternal protein increases while that of paternal protein decreases.  

1. What can we not conclude?

Whether or not the decrease in paternal protein intensity CAUSES increase in maternal intensity

1. What can we infer from the data?

Imprinting of ube3a at the paternal allele is initiated upon neuronal differentiation. Increase in Ube3a from the maternal allele compensates for loss of expression of the paternal allele in neurons.

Figure 3: Ube3a protein levels are higher in mouse and opossum despite no imprinting in opossum

 

1. What does the several SNV mean?

1. Single nucleotide variant

1. What experiments did they do? What are the controls and why did they have that?

 

1. What was the purpose of these experiments? What question were they addressing?
2. What do the data show?

1. The alleleic ratio of maternal and paternal in UBE3A (opossum) are similar in the SNVs of the cortex. However, the number for paternal is slightly higher in SNV4 than maternal though not statistically significant.
2. Expression of UBE3A in CNS is significantly higher than in N-CNS

1. What can we conclude from the data?

1. UBE3A is biallelically expressed in the opossum
2. Imprinting is NOT NECESSARY to result in high UBE3A expression in CNS compared to N-CNS

1. What can we not conclude?

1. What can we say about slightly higher ratio in paternal?
2. If imprinting is not the cause for higher UBE3a expression in imprinted mouse  

1. What can we infer from the data?

1. b) There are other mechanisms that induce expression of mouse UBE3A particularly in the CNS compared to N-CNS

1.     C) imprinting in the mouse may cause higher expression of Ube3a in the cortex compared to expression of UBE3A in opossum.