Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Jan 10:13:77.
doi: 10.3389/fncir.2019.00077. eCollection 2019.

Multiplex Neural Circuit Tracing With G-Deleted Rabies Viral Vectors

Affiliations

Multiplex Neural Circuit Tracing With G-Deleted Rabies Viral Vectors

Toshiaki Suzuki et al. Front Neural Circuits. .

Abstract

Neural circuits interconnect to organize large-scale networks that generate perception, cognition, memory, and behavior. Information in the nervous system is processed both through parallel, independent circuits and through intermixing circuits. Analyzing the interaction between circuits is particularly indispensable for elucidating how the brain functions. Monosynaptic circuit tracing with glycoprotein (G) gene-deleted rabies viral vectors (RVΔG) comprises a powerful approach for studying the structure and function of neural circuits. Pseudotyping of RVΔG with the foreign envelope EnvA permits expression of transgenes such as fluorescent proteins, genetically-encoded sensors, or optogenetic tools in cells expressing TVA, a cognate receptor for EnvA. Trans-complementation with rabies virus glycoproteins (RV-G) enables trans-synaptic labeling of input neurons directly connected to the starter neurons expressing both TVA and RV-G. However, it remains challenging to simultaneously map neuronal connections from multiple cell populations and their interactions between intermixing circuits solely with the EnvA/TVA-mediated RV tracing system in a single animal. To overcome this limitation, here, we multiplexed RVΔG circuit tracing by optimizing distinct viral envelopes (oEnvX) and their corresponding receptors (oTVX). Based on the EnvB/TVB and EnvE/DR46-TVB systems derived from the avian sarcoma leukosis virus (ASLV), we developed optimized TVB receptors with lower or higher affinity (oTVB-L or oTVB-H) and the chimeric envelope oEnvB, as well as an optimized TVE receptor with higher affinity (oTVE-H) and its chimeric envelope oEnvE. We demonstrated independence of RVΔG infection between the oEnvA/oTVA, oEnvB/oTVB, and oEnvE/oTVE systems and in vivo proof-of-concept for multiplex circuit tracing from two distinct classes of layer 5 neurons targeting either other cortical or subcortical areas. We also successfully labeled common input of the lateral geniculate nucleus to both cortico-cortical layer 5 neurons and inhibitory neurons of the mouse V1 with multiplex RVΔG tracing. These oEnvA/oTVA, oEnvB/oTVB, and oEnvE/oTVE systems allow for differential labeling of distinct circuits to uncover the mechanisms underlying parallel processing through independent circuits and integrated processing through interaction between circuits in the brain.

Keywords: anatomy; multiplex; neural circuit; rabies virus; transsynaptic targeting.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Design and function of avian sarcoma leukosis virus (ASLV)/RV-G chimeric envelope-pseudotyped RVΔG. (A) Depiction of oEnvX-pseudotyped RVΔG production. 1st step: recovery of RVΔG viral particles from DNA plasmids and amplification of recovered RVΔG in B7GG cells. 2nd step: generation of RV pseudotyped with oEnvX from oEnvX-expressing cells that were infected with native RVΔG. 3rd step: evaluation of generated viral vectors. The viral supernatant was filtrated for the in vitro assay and further concentrated by ultra-centrifugation for the in vivo assay. (B) Design of the ASLV envelope/RV-G chimeric envelope. oEnvA, oEnvB, and oEnvE were composed of the extracellular and transmembrane domains of EnvA, EnvB, and EnvE envelopes, respectively, and the intracellular domain of RV-G. (C) Specific infection of mixed oEnvA-RVΔG-tagBFP, oEnvB-RVΔG-DsRed, and oEnvE-RVΔG-GFP to HEK293t cells expressing their respective receptors TVA800, TVBS3, and DR46-TVB. Signals were derived from native fluorescence. Scale bar: 50 μm.
Figure 2
Figure 2
Generation of an optimized oEnvA/oTVA system for unambiguously marking starter cells. (A) Design of TVA variants fused to a fluorescent protein. TVA950 and its lower-affinity version TVA950E66T (oTVA-L) were fused to mCherry or iRFP670. (B) Membrane localization of oTVA-L-mCherry (left) and oTVA-L-iRFP (right) in transfected HEK293t cells. Scale bar: 20 μm. (C) Infection efficiency of oEnvA-RVΔG to cells expressing TVA950-mCherry, oTVA-L-mCherry, or oTVA-L-iRFP. Each column represents the mean ± SEM (n = 3). **P < 0.01 vs. TVA950-mCherry (Tukey’s multiple tests). n.s.: not significant. n.d.: not detected. (D) In vivo evaluation of oTVA-L-iRFP function in the mouse brain. AAV2/9-DIO-oTVA-L-mCherry or AAV2/9-DIO-oTVA-L-iRFP was injected to the V1 of layer 5-specific Tlx3-Cre mice to introduce oTVA-L-mCherry or oTVA-L-iRFP in layer 5 neurons. oEnvA-RVΔG-GFP was injected at the location of the AAV injection. (E) Reliable targeting of oEnvA-RVΔG-GFP to oTVA-L-mCherry-expressing cells. oTVA-L-mCherry was expressed in layer 5 neurons in the V1 of Tlx3-Cre mice. All oEnvA-RVΔG-GFP-infected neurons expressed oTVA-L-mCherry. Signals were derived from native fluorescence. Scale bar: 100 μm. The inset shows high magnifications of areas indicated by the white square. Scale bar: 20 μm. (F) Reliable targeting of oEnvA-RVΔG-GFP to oTVA-L-iRFP670-expressing cells. oTVA-L-iRFP was expressed in layer 5 neurons in the V1 of Tlx3-Cre mice. All oEnvA-RVΔG-GFP-infected neurons expressed oTVA-L-iRFP. Signals were derived from native fluorescence. Scale bar: 100 μm. The inset shows high magnifications of areas indicated by the white square. Scale bar: 20 μm.
Figure 3
Figure 3
Generation of an optimized oEnvB/oTVB system for viral targeting. (A) Design of TVB variants fused to a blue fluorescent protein. TVBS3ΔDD-BFP was deficient in the death domain from TVBS3 and fused to tagBFP. TVBS3ΔDD-TVA-BFP (oTVB-H) has a TVA950 intracellular region instead of the TVBS3 intracellular region and fused-tagBFP. (B) Subcellular localization of TVBS3-BFP (left) and oTVB-H-BFP (right) in transfected-HEK293t cells. Abnormal localization of TVBS3-BFP in the cell membrane and cytoplasmic region. Arrowhead: aggregation of TVBS3-BFP. Membrane localization of oTVB-H-BFP. Scale bar: 20 μm. (C) Infectious efficiency of oEnvB-RVΔG to cells expressing TVBS3ΔDD, TVBS3-BFP, or oTVB-H-BFP.Each column represents the mean ± SEM (n = 3). *P < 0.05 vs. oTVB-H (Tukey’s multiple tests). n.d.; not detected. (D) In vivo evaluation of oTVB-H function in the mouse brain. AAV2/9-DIO-oTVB-H-BFP was injected to the V1 of Tlx3-Cre mice to introduce oTVB-H in layer 5 neurons. oEnvA-RVΔG-DsRed was injected at the location of the AAV injection. (E) Evaluation of in vivo function of TVBS3-TVA-BFP (oTVB-H) in Tlx3-Cre mice. oEnvB-RVΔG-DsRed infected layer 5 neurons without visible oTVB-H-tagBFP signals, suggesting that oTVB-H was sufficiently sensitive for oEnvB-RVΔG-DsRed to infect cells expressing oTVB-H at a low expression level. The arrowhead indicates non-specific infection to neurons in layer 2/3. Signals were derived from native fluorescence. Scale bar: 100 μm. The inset shows high magnifications of areas indicated by the white square. Scale bar: 20 μm. (F) Generation of TVB mutants fused with a blue fluorescent protein. TVBS1 has one mutation (S62C). TVBr2 has two mutations (S62C and C125S). TVBS3ΔCRD2 is devoid of CRD2 from TVBS3. TVBS3ΔCRD3 is devoid of CRD3 from TVBS3. (G,H) Infection efficiency of (G) oEnvB-RVΔG and (H) oEnvE-RVΔG to cells expressing one of the tagBFP-fused TVB mutants. Each column represents the mean ± SEM (n = 3). *P < 0.05, ***P < 0.001 vs. TVBS3-TVA-BFP (oTVB-H) (Dunnett’s multiple test). (I) Evaluation of in vivo function of TVBS3ΔCRD3-TVA-BFP (oTVB-L). oEnvB-RVΔG-DsRed infected oTVB-L-BFP-expressing neurons of layer 5-specific Tlx3-Cre mice. Signals were derived from native fluorescence. Scale bar: 100 μm. The inset shows high magnifications of areas indicated by the white square. Scale bar: 20 μm.
Figure 4
Figure 4
Generation of an optimized oEnvE/oTVE system for viral targeting. (A) Design of DR46-TVB variants fused to a red or blue fluorescent protein. DR46-TVB-TVA includes the extracellular-transmembrane region of DR46-TVB and the intracellular region of TVA950, according to the design of oTVB (see Figures 3A,B). (B) Membrane localization of DR46-TVB-TVA (oTVE-H)-BFP in HEK293t cells. The TagBFP signal was observed in the cell membrane. Signals were derived from native fluorescence. Scale bar: 20 μm. (C) Infection efficiency of oEnvE-RVΔG to cells expressing DR46-TVB or oTVE-H. Each column represents the mean ± SEM (n = 3). **P < 0.01 vs. DR46-TVB, ##P < 0.01 vs. oTVE-H (Tukey’s multiple tests). (D) Evaluation of in vivo function of oTVE-H-mCherry in Tlx3-Cre mice. Typical images of oEnvE-RVΔG-GFP-infected layer 5 neurons although the layer 5 neurons do not have visible expression of oTVE-H-mCherry. These data suggest that oTVE-H was sufficiently sensitive for oEnvE-RVΔG-GFP to infect layer 5 neurons expressing oTVE-H at a low expression level. Scale bar: 100 μm. The inset shows high magnifications of areas indicated by the white square. Signals were derived from native fluorescence. Scale bar: 20 μm. (E) The 2A-mediated oTVE-H expression system. oTVE-H was introduced by the H2B-tagBFP-P2A-oTVE-H cassette. oEnvE-RVΔG-GFP infected neurons expressing both nuclear H2B-tagBFP and oTVE-H in Tlx3-Cre mice. The arrowhead indicates non-specific infection to neurons in layers 2/3. Signals were derived from native fluorescence. Scale bar: 100 μm. The inset shows high magnifications of areas indicated by the white square. Scale bar: 20 μm. (F) The IRES-mediated the oTVE expression system for reliable detection of oTVE-H-expressing cells. oTVE-H was expressed downstream of the IRES sequence. oEnvE-RVΔG-GFP infected neurons expressing visible H2B-mRuby3 as well as oTVE-H in Tlx3-Cre mice. oEnvE-RVΔG-GFP-infected neurons expressed nuclear mRuby3. Signals were derived from native fluorescence. Scale bar: 100 μm. The inset shows high magnifications of areas indicated by the white square. Scale bar: 20 μm.
Figure 5
Figure 5
Independence of the oEnvA/oTVA and oEnvB/oTVB systems. (A) In vitro evaluation of specificity of oEnvA-RVΔG-GFP and oEnvB-RVΔG-DsRed to cells expressing either oTVA-L-iRFP or oTVB-L-BFP. oEnvA-RV and oEnvB-RV mixed viruses applied to HEK293t cells expressing oTVA-L-iRFP or oTVB-L-BFP transitorily. (B) No cross-infectivity of oEnvA-RVΔG-GFP and oEnvB-RVΔG-DsRed to cells expressing either oTVA-L-iRFP or oTVB-L-BFP. Evaluation of in vitro infectious specificity. Each column represents the mean ± SEM (n = 3). n.d.: not detected. (C) In vivo infectious specificity of oTVA-L-iRFP and oTVB-L-BFP that are expressed in two different locations of the brain. AAV-CAG-DIO-oTVA-L-iRFP and AAV-CaMKIIa-DIO-oTVB-L-BFP were injected in two different locations of Tlx3-Cre mice, 800 μm apart. A mixture of oEnvA-RVΔG-DsRed and oEnvB-RVΔG-GFP was injected between the two injection sites of AAV. (D) Non-co-infection with oEnvA-RVΔG and oEnvB-RVΔG. Signals were derived from native fluorescence. Scale bar: 100 μm. (E) In vivo infection specificity of oTVA-L-iRFP and oTVB-L-BFP that are expressed in two different cell types. oTVA-L-iRFP and oTVB-L-BFP are exclusively expressed by a distinct promotor and a transgenic Cre line. Inhibitory neurons were targeted by the inhibitory neuron-specific mDlx promoter and layer 5 neurons recombined the AAV genome to excise the flox cassette and eliminate gene expressions. Excitatory neurons in layer 5 of the cortex were targeted by a combination of the excitatory neuron-specific CaMKIIa promoter and the layer 5-specific Tlx3-Cre line. AAV-mDlx-oTVA-L-iRFP and AAV-CaMKIIa-DIO-oTVB-L-BFP were co-injected into the V1 of Tlx3-Cre mice. oEnvA-RVΔG-DsRed and oEnvB-RVΔG-GFP were co-injected into the location of the AAV injection. (F) No cross-infectivity between oEnvA-RVΔG and oEnvB-RVΔG. Signals were derived from native fluorescence. Scale bar: 100 μm.
Figure 6
Figure 6
Multiplex circuit tracing with the oEnvA/oTVA and oEnvB/oTVB systems. (A) Multiplex monosynaptic tracing of two distinct classes of layer 5 neurons targeting either other cortical or subcortical areas. Retrograde AAVs were injected into two different locations of layer 5-specifc Tlx3-Cre mice: AAV2retro-CBh-DIO-oTVA-L-iRFP-P2A-oG and AAV2retro-CBh-oTVB-L-BFP-P2A-oG in the lateral area of the higher visual cortex (V2L) and superior colliculus (SC), respectively. V2L-projecting V1 neurons in layer 5 expressed both oTVA-L-iRFP and oG while SC-projecting V1 neurons expressed both oTVB-L-BFP and oG. After the AAV injections, oEnvA-RVΔG-DsRed and oEnvB-RVΔG-GFP were co-injected into the V1. (B) Coronal brain sections illustrating anatomical regions shown in (CI; adapted from The Mouse Brain in Stereotaxic Coordinates, 3rd edition, Paxinos and Franklin, ; G). (C) Viral targeting to two distinct classes of layer 5 neurons depending on projection targets. oTVA-L-iRFP-expressing starter neurons were specifically infected with oEnvA-RVΔG-DsRed, whereas oTVB-L-BFP-expressing starter neurons were infected with oEnvB-RVΔG-GFP. DsRed and GFP signals were enhanced by immunostaining. Closed arrowheads indicate oTVA-L-iRFP+/oEnvA- RVΔG-DsRed + starter neurons. Opened arrowheads indicate oTVB-L-BFP+/oEnvB-RVΔG-GFP+ starter neurons. Scale bar: 100 μm. (D–I) Distributions of presynaptic neurons connected to two distinct classes of layer 5 neurons. Neurons trans-synaptically infected with RVΔG-DsRed and RVΔG-GFP were distributed in the V1, V2L, and RSD (D), V1, V2L, V2ML, V2MM, and RSD (E), dLGN (F), dLGN and LP (G), LPtA, MPtA, and RSD (H), AM, AVDM, and AVVL (I). DsRed and GFP signals were enhanced by immunostaining. Scale bar: 200 μm. Abbreviations: AM, anteromedial thalamic nucleus; AVDM, anterovent thalamic nucleus, dorsomedial part; AVVL, anteroventral thalamic nucleus, ventrolateral part; RSD, retrosplenial dysgranular cortex; dLGN, dorsal lateral geniculate nucleus; LP, lateral parietal association cortex; LPtA, medial parietal association cortex; MPtA, lateral posterior thalamic nucleus; V2L, secondary visual cortex, lateral area; V2ML, secondary visual cortex, mediolateral area; V2MM, secondary visual cortex, mediomedial area.
Figure 7
Figure 7
Labeling of common input with multiplex circuit tracing. (A) Multiplex monosynaptic tracing of two distinct classes of layer 5 excitatory or inhibitory neurons. Mixed AAVs were injected into the V1 of layer 5-specifc Tlx3-Cre mice: AAV2/2-mDlx-flox-oTVA-L-iRFP, AAV2/2-CaMKIIa-DIO-oTVB-L-BFP, and AAV2/9-CAG-H2B-tagBFP-P2A-oG. After the AAV injection, oEnvA-RVΔG-DsRed and oEnvB-RVΔG-GFP were co-injected into the same location of the V1. (B) Viral targeting to two distinct classes of V1 neurons depending on cell types. oTVA-L-iRFP-expressing inhibitory neurons were specifically infected with oEnvA-RVΔG-DsRed, whereas oTVB-L-BFP-expressing excitatory neurons in layer 5 were infected with oEnvB-RVΔG-GFP. The rabies glycoprotein oG was expressed in cells positive for nuclear tagBFP. Closed arrowheads indicate presynaptic neurons co-infected with both oEnvA-RVΔG-DsRed and oEnvB-RVΔG-GFP. DsRed and GFP signals were enhanced by immunostaining. Scale bar: 200 μm. (C–E) Low and high magnification images of dLGN neurons monosynaptically connected to two distinct classes of V1 neurons. DsRed and GFP signals were enhanced by immunostaining. Closed arrowheads indicate presynaptic neurons co-infected with both oEnvA-RVΔG-DsRed and oEnvB-RVΔG-GFP. Scale bar of overview: 500 μm. Scale bar of dLGN view: 100 μm. Coronal brain sections illustrating anatomical regions shown in (B–E; adapted from The Mouse Brain in Stereotaxic Coordinates, 3rd edition, Paxinos and Franklin, 2008).
Figure 8
Figure 8
Multiplex circuit analysis with rabies viral vectors. Simultaneous RV trans-synaptic tracing can be achieved by designing three steps. (1) Select viral envelopes and their receptors. oEnvX/oTVX systems developed in the present study can be used for rabies viral targeting. (2) Select cells of interest. Cell-type-specific promoters, connection-mediated gene transfer, electroporation, and transgenic animals are available to target particular populations. (3) Select genes of interest. Fluorescent proteins, imaging probes, optogenetic/chemogenetic tools, and recombinases can be expressed from the rabies viral vectors.

Similar articles

Cited by

References

    1. Abdelfattah A. S., Kawashima T., Singh A., Novak O., Liu H., Shuai Y., et al. . (2019). Bright and photostable chemigenetic indicators for extended in vivo voltage imaging. Science 365, 699–704. 10.1126/science.aav6416 - DOI - PubMed
    1. Adkins H. B., Blacklow S. C., Young J. A. T. (2001). Two functionally distinct forms of a retroviral receptor explain the nonreciprocal receptor interference among subgroups B, D, and E avian leukosis viruses. J. Virol. 75, 3520–3526. 10.1128/JVI.75.8.3520-3526.2001 - DOI - PMC - PubMed
    1. Adkins H. B., Brojatsch J., Naughton J., Rolls M. M., Pesola J. M., Young J. A. T. (1997). Identification of a cellular receptor for subgroup E avian leukosis virus. Proc. Natl. Acad. Sci. U S A 94, 11617–11622. 10.1073/pnas.94.21.11617 - DOI - PMC - PubMed
    1. Adkins H. B., Brojatsch J., Young J. A. T. (2000). Identification and characterization of a shared TNFR-related receptor for subgroup B, D, and E avian leukosis viruses reveal cysteine residues required specifically for subgroup E viral entry. J. Virol. 74, 3572–3578. 10.1128/jvi.74.8.3572-3578.2000 - DOI - PMC - PubMed
    1. Alexander G. M., Rogan S. C., Abbas A. I., Armbruster B. N., Pei Y., Allen J. A., et al. . (2009). Remote control of neuronal activity in transgenic mice expressing evolved G protein-coupled receptors. Neuron 63, 27–39. 10.1016/j.neuron.2009.06.014 - DOI - PMC - PubMed

Publication types

MeSH terms