Whole brain mapping of inputs

to inhibitory interneurons

in prefrontal cortex

Yang Xuan
Carlén Lab

Networks for Networks
28th Nov 2014

yang.xuan@ki.se

Topics

Preliminary findings SOM and PV.

• Regions of interest:
  • Basal forebrain
  • Thalamus.
  • Septum.
  • Amygdala.
  • Hippocampus.

• Background and aims.
• Dual versus single rabies vector systems.
• Whole-brain reconstruction.
• Local versus long-range inputs

Background and Aims

• Retrograde tracing
• SAD B19G2
• Five genes

Rabies virus

Rabies-virus-glycoprotein (RVG) necessary and sufficient for transsynaptic spread.

• Avian EnvA-TVA
• Delivering TVA and RVG into starting cells

Cell-type-specific delivery

Dual vector system (traditional)

• Starter cells definition
  • True positive:
    • + TVA-mCherry
    • + Rabies-EGFP.
    • + RVG
  • False positives.
    • + TVA-mCherry
    • + Rabies-EGFP.
    • -  RVG

1. Overestimation of starting population
2. Not possible to dissect local input (e.g. between layers)

Leads to the following problems:

Pollak et al. 2014, Wall et al. 2013, Uchida et al. 2012 

Dual vector system

• Tag RVG with fluorescent protein (e.g. EF1a-DIO-RVG-Venus).

        Problem: RVG is sensitive to tag, miss-folding (pers. comm. Ed Callaway).

• All in one vector (e.g. EF1a-DIO-TVA-mCherry-t2A-RVG).

        Problem: To large to fit in standard vectors like AAV and lentivirus.

• One vector without tag (e.g. EF1a-DIO-TVA-t2A-RVG).

        Problem: 

• Single vector rabies system. (e.g. EF1a-DIO-TVA-V5-t2A-RVG)​

SOLUTION

Single vector system

V5   Paramyxovirus SV5 

a small epitope present on the P and V proteins of SV5.

• 14 amino acids       42 bp.
• Works good with AAV vectors.
• Fusion protein with TVA on C-terminal        membrane bound.

Single vector system

I've also extended the single vector rabies virus system with additional rabies varieties and flavors.
 

• Tracing:  3'-N-P-M-EGFP-L-5'.
• Gain of function: 3'-N-P-M-ChR2-mCherry-L-5'.
• Loss of function: 3'-N-P-M-Jaws-EGFP-L-5'.
• Input sensitivity: 3'-N-P-M-SSFO-EYFP-L-5'.

Disadvantage: toxicity of the rabies virus itself. Mechanism unknown. 
Puts a limited time frame from injection to end-point of experiment.

Whole-Brain Reconstruction

Pollak et al. 2014

Reconstructing brain from sectioned tissue

WholeBrain Web interface

similar to...

works with...

Local versus long-range input

• How much local versus long-range input.
• median spherical radial distance

Local versus long-range input

Local versus long-range input

• No marked difference between            and      .
• Statistical power issue.

This severely limits any quantitative comparison across all unique brain regions that cells were found in.

Until sample size is increased          Qualitative findings

SOM

PV

Cortical input

PV-cre mouse ID# 195

SOM-cre mouse ID# 80

~80% of all inputs are cortical input

Hippocampus

• ChAT
• Rabies-EGFP

Basal forebrain

• FOXP2
• PV

SOM-cre

PV-cre

Medial Septum

SOM-cre

PV-cre

Rabies-EGFP

CAhT

Amygdala

Amygdala Immunohistochemistry

Rabies-EGFP

CAhT

FoxP2

SOM-cre

PV-cre

• Ventral group of the dorsal thalamus

• Summary

• Bullet One
• Bullet Two
• Bullet Three

• On-going work

• Tracing project:
• Other interneurons (VIP)
• Principal cells (Coinjection with CaMKIIa-CRE)
• New vector designs: CaMKIIa-TVA-V5-t2A-RG

• Acknowledgement

• Bullet One
• Bullet Two
• Bullet Three

• Thank you!

•     Advantages and Disadvantages

• TVA-V5-RG Vs TVA-mCherry+RG(TVA-mCherry/non-RG infected cells)
• Local input
• •Disadvantages:
• Staining

• Alternative:
  
•   AAV8/synP-FLEX-sTpEpB
•   (Keigo Kohara et al., Nature Neuroscience 2014)
  
•   Reduction of total genome length by:
  
•     using smaller promotor(human synapsin-1 promotor)
•     using smaller polyadenylation signal(bovine growth hormone polyadenylation site)
  
•     removing all extraneous sequences and restriction sites
  
•     7 days+ 7 days