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
[Preprint]. 2023 Jul 25:2023.07.21.550095.
doi: 10.1101/2023.07.21.550095.

Exuberant de novo dendritic spine growth in mature neurons

Sarah Krüssel  1 , Ishana Deb  1 , Seungkyu Son  2 , Gabrielle Ewall  3 , Minhyeok Chang  1 , Hey-Kyoung Lee  3   4 , Won do Heo  2 , Hyung-Bae Kwon  1
Affiliations

Exuberant de novo dendritic spine growth in mature neurons

Sarah Krüssel et al. bioRxiv. .

Update in

Abstract

Dendritic spines are structural correlates of excitatory synapses maintaining stable synaptic communications. However, this strong spine-synapse relationship was mainly characterized in excitatory pyramidal neurons (PyNs), raising a possibility that inferring synaptic density from dendritic spine number may not be universally applied to all neuronal types. Here we found that the ectopic expression of H-Ras increased dendritic spine numbers regardless of cortical cell types such as layer 2/3 pyramidal neurons (PyNs), parvalbumin (PV)- and vasoactive intestinal peptide (VIP)-positive interneurons (INs) in the primary motor cortex (M1). The probability of detecting dendritic spines was positively correlated with the magnitude of H-Ras activity, suggesting elevated local H-Ras activity is involved in the process of dendritic spine formation. H-Ras overexpression caused high spine turnover rate via adding more spines rather than eliminating them. Two-photon photolysis of glutamate triggered de novo dendritic spine formation in mature neurons, suggesting H-Ras induced spine formation is not restricted to the early development. In PyNs and PV-INs, but not VIP-INs, we observed a shift in average spine neck length towards longer filopodia-like phenotypes. The portion of dendritic spines lacking key excitatory synaptic proteins were significantly increased in H-Ras transfected neurons, suggesting that these increased spines have other distinct functions. High spine density caused by H-Ras did not result in change in the frequency or the amplitude of miniature excitatory postsynaptic currents (mEPSCs). Thus, our results propose that dendritic spines possess more multifaceted functions beyond the morphological proxy of excitatory synapse.

Keywords: Dendritic spines; H-Ras; Interneurons; Spine formation; Synaptogenesis.

PubMed Disclaimer

Conflict of interest statement

DECLARATION OF INTEREST The authors declare no conflict of interest.

Figures

Figure 1:
Figure 1:. H-Ras increases spine number in cortical pyramidal neurons.
(A) Left: Schematic depiction of H-Ras sensor constructs and its controls (RBDRAF, ΔH-Ras, and Control). Right: Schematic of the mode of action of ddFP-based H-Ras sensor. (B) Virus injection scheme and experimental timeline. (C) Two-photon microscopy images showing representative dendrites (apical distal, apical proximal, and basal) of pyramidal neurons expressing Flex-tdTomato in the primary motor cortex of a typical acute brain slice made from C57Bl6 mice at ~P60 injected 4-weeks prior with CaMKII-Cre, Flex-tdTomato and either Flex-pCAG-B3-RafRBD-2A-GA-HRas (H-Ras), Flex-pCAG-B3-RafRBD-2A-GA (RBDRAF) or Flex-pCAG-B3–2A-GA (ΔH-Ras). Scale bar, 2 μm. (D) A summary graph showing the spine density. Dots represent average spine number from each neuron (dots) and bars indicate mean ± SEM, respectively for each condition (Control: 0.7259 ± 0.05162, N=17; H-Ras:1.0316 ± 0.0743, N=15; ΔH-Ras: 0.73881 ± 0.10413, N=10; and RBDRAF: 0.6554 ± 0.07848, N=7). The dotted line represents the average spine density of the tdTomato only control. *p<0.05 (Two-way ANOVA) (E) Separation of spine density analysis into the three dendritic locations: apical distal, apical proximal and basal. Apical distal: Control (0.931 ± 0.0776), H-Ras (1.06549 ± 0.09451), ΔH-Ras (0.79241 ± 0.14282), and RBDRAF (0.77855 ± 0.09053); Apical proximal: Control (0.37316 ± 0.0448), H-Ras (0.84397 ± 0.0987), ΔH-Ras (0.30431 ± 0.04335), and RBDRAF (0.35444 ± 0.06879); basal: Control (0.5743 ± 0.05578), H-Ras (0.95856 ± 0.06823), ΔH-Ras (0.58031 ± 0.09471), and RBDRAF (0.5017 ± 0.107). Control: N=17; H-Ras: N=15; ΔH-Ras: N=10; RBDRAF: N=7. *p<0.05, **p<0.01, ***p<0.001 (Two-way ANOVA).
Figure 2:
Figure 2:. Spine density changes by H-Ras ectopic expression is independent of cell types.
(A) Virus injection scheme and experimental timeline. (B) Representative two-photon microscopy images of parvalbumin-positive interneurons (PV-INs). Scale bar: 10 μm. (C) Two-photon microscopy images showing representative dendrites (proximal and distal) of PV-INs expressing Flex-tdTomato and either Flex-pCAG-B3-RafRBD-2A-GA-HRas (H-Ras), Flex-pCAG-B3-RafRBD-2A-GA (RBDRAF) or Flex-pCAG-B3–2A-GA (ΔH-Ras). Scale bar, 2 μm. (D) A superimposed bar and dot graph showing the spine density of each neuron (dot) and mean ± SEM (bar), respectively, for each condition (Control, H-Ras, RBDRAF, and ΔH-Ras). (Top) Spine density averaged from all imaged dendritic branches; Control (0.17519 ± 0.02112); H-Ras (0.61928 ± 0.06211); ΔH-Ras (0.18324 ± 0.01037); RBDRAF (0.20159 ± 0.01719); (Middle) spine density of proximal dendrites; Control (0.15556 ± 0.01494); H-Ras (0.45558 ± 0.04228); ΔH-Ras (0.16542 ± 0.01075); RBDRAF (0.18838 ± 0.01431); (Bottom) spine density of distal dendrites; Control (0.19191 ± 0.02918); H-Ras (0.71365 ± 0.07629); ΔH-Ras (0.21208 ± 0.02675); RBDRAF (0.25007 ±. 0.03138). The dotted line represents the average spine density of dendrites expressing tdTomato only (Control). Neurons expressing H-Ras showed a higher rate of dendritic spines throughout all dendritic regions. Control: N=16; H-Ras: N=22; ΔH-Ras: N=14; RBDRAF: N=13. ***p<0.001 (Kruskal Wal Anova). (E) Representative two-photon images of vasoactive intestinal peptide-expressing interneurons (VIP-INs). Scale bar: 10 μm. (F) Two-photon microscopy images showing representative dendrites (apical distal, apical proximal, and basal) of VIP INs expressing Flex-tdTomato in the primary motor cortex of a typical acute brain slice made from VIP-Cre mice at ~P60 injected 4-weeks prior with Flex-tdTomato and either H-Ras, or ΔH-Ras. Scale bar, 2 μm. (G) A superimposed bar and dot graph showing the spine density of each neuron (dots) and mean ± SEM (bar), respectively, for each condition (H-Ras, and ΔH-Ras). (left) Spine density averaged from all imaged dendritic branches; H-Ras (0.43456 ± 0.04397); ΔH-Ras (0.13683 ± 0.13375); (middle) spine density of proximal dendrites; H-Ras (0.42397 ± 0.0354); ΔH-Ras (0.14144 ± 0.0175); (right) spine density of distal dendrites; H-Ras (0.43697 ± 0.04737); ΔH-Ras (0.13057 ± 0.01376). The dotted line represents the average spine density of dendrites expressing ΔH-Ras. Neurons expressing H-Ras showed a higher spine density throughout all dendritic regions. H-Ras: N=20; ΔH-Ras: N=12; ***p<0.001 (Two Sample T-Test).
Figure 3:
Figure 3:. Local H-Ras activity facilitates de novo growth of spines.
(A) Schematic of organotypic slice preparation and timeline. (B) Representative dendrite images expressing tdTomato (top) and H-Ras (tdTomato + pCAG-B3-RafRBD-2A-GA-HRas) (bottom) collected via a two-photon microscope at P12. DNA was biolistically transfected to organotypic cortical slices at P1. Scale bar: 2 μm. (C) A summary graph showing the spine density of each analyzed dendritic section (~30 μm) (dots) and the mean spine density ± SEM (bar) of Control (0.49719 ± 0.02549) and H-Ras neurons (0.6445 ± 0.02608). The dotted line represents the average spine density of Control dendrites. Control: N= 53, H-Ras: N=52; ***p<0.001 (Two Sample T-Test). (D) tdTomato and H-Ras biosensor expression in a representative organotypic pyramidal neuron: red= tdTomato, green= H-Ras biosensor. Scale bar: (E) Magnification of dendritic branch shown in D (white box) (top) tdTomato, (middle) H-Ras biosensor GFP, and (bottom) merge. Colored arrow heads indicate examples of high H-Ras activity (orange) or low H-Ras activity (light blue). Scale bar: 2 μm. (F) Scatter plot showing that higher H-Ras activity (relative green/red ratio) correlates with a higher percentage of detecting a spine in close proximity (1 μm). The sigmoidal curve fitting to the data shows an R-Value of 0.87415. (G) Correlation matrix comparing H-Ras activity values with the existence of a spine in a 1 μm radius; (left) experimental data, (right) shuffled data control. H-Ras activity correlates to a higher extent with the existence of dendritic spines in the experimental data in comparison to shuffled data. (H) Virus injection scheme and experimental timeline. (I) Example images of high frequency glutamate uncaging (HFU) experiments (white circles, 40 pulses at 10Hz) on adult pyramidal neurons (~P60): (top left) ΔH-Ras + MNI glutamate, (top right) H-Ras + MNI glutamate, (bottom left) ΔH-Ras without MNI-glutamate, and (bottom right) H-Ras without MNI-glutamate. All neurons expressed tdTomato as cell markers. White arrowheads indicate de novo spine formation after HFU. Scale bar, 2 μm. (J) Success rate of de novo spine formation by HFU at P60 in pyramidal neurons expressing tdTomato and either ΔH-Ras (19.2%) or H-Ras (52%). In the absence of MNI-glutamate (mock) neither ΔH-Ras nor H-Ras exhibited any de novo spine formation. ΔH-Ras: N= 24 trials, 16 cells; H-Ras: N= 26 trials, 15 cells; ΔH-Ras mock: N= 4 trials, 4 cells; H-Ras mock: N= 8 trials, 7 cells. ***p<0.001 (Chi-Square Test).
Figure 4:
Figure 4:. Increased spine number does not represent features of functional excitatory synapses.
(A) Virus injection scheme and experimental timeline. (B) Light microscopy images of layer 2/3 pyramidal neuron recorded in whole-cell voltage-clamp-mode in acute slices: whole motor cortical brain region (left), magnified image: light microscopy and fluorescent (right bottom), schematic of pyramidal neurons being fluorescent positive (+) or fluorescent negative (−) (right top). (C) Examples traces of mEPSCs in ΔH-Ras+ (blue), H-Ras+ (orange), ΔH-Ras− (grey), and H-Ras− neurons (black). (D-G) mEPSC frequencies (D), amplitudes (E), rise times (F), and decay times (G) in ΔH-Ras+ (blue), H-Ras+ (orange), ΔH-Ras− (grey), and H-Ras− neurons (black). Each dot represents an average value from one neuron, and the dotted line represents mean ± SEM. (H) Example confocal images of ΔH-Ras (left) and H-Ras of pyramidal neurons and magnified dendrites (right). From top to bottom: image of pyramidal neuron expressing tdTomato, postsynaptic protein Homer1 stained with Alexa Fluor 633 (green), presynaptic protein bassoon stained with Alexa Fluor 405 (blue), merge of all channels. Scale bars: 10 μm (pyramidal neuron), 3 μm (dendrite). (I) Magnification image of (H) visualizing single spines of ΔH-Ras (left) and H-Ras (right). From top to bottom: postsynaptic protein Homer1 (green) stained with Alexa Fluor 633 with and without dendrite visualization, presynaptic protein bassoon with Alexa Fluor 405 (blue) stained with Alexa Fluor 405 with and without dendrite visualization, merge of channels with and without dendrite visualization. Scale bar: 0.75 μm. (J) Imaris reconstructed image of dendrites and dendritic spines in conjunction with Bassoon reconstruction (blue) or Homer1 reconstruction (green) of ΔH-Ras (left) and H-Ras neurons (right). Scale bar: 0.75 μm. (K) Imaris reconstruction of dendrites/spines together with Homer1 (green, top) or bassoon (blue, bottom) of ΔH-Ras (left) and H-Ras neurons (right). (L-M) Analysis of fraction of spines encompassing Homer1 puncta (L) or having close contact (0.1 μm) to bassoon puncta (M) in either ΔH-Ras or H-Ras neurons shows that neurons expressing ectopic H-Ras have reduced number of spines expressing Homer1 as well as reduced bassoon contacts. ΔH-Ras: N= 12, H-Ras: N= 9; **p<0.01, ***p<0.001 (Two-Sample Kolmogorov-Smirnov Test).
See this image and copyright information in PMC

References

    1. Fu M., Yu X., Lu J., and Zuo Y. (2012). Repetitive motor learning induces coordinated formation of clustered dendritic spines in vivo. Nature 2012 483:7387 483, 92–95. 10.1038/nature10844. - DOI - PMC - PubMed
    1. Grutzendler J., Kasthuri N., and Gan W.B. (2002). Long-term dendritic spine stability in the adult cortex. Nature 2003 420:6917 420, 812–816. 10.1038/nature01276. - DOI - PubMed
    1. Holtmaat A.J.G.D., Trachtenberg J.T., Wilbrecht L., Shepherd G.M., Zhang X., Knott G.W., and Svoboda K. (2005). Transient and persistent dendritic spines in the neocortex in vivo. Neuron 45, 279–291. 10.1016/j.neuron.2005.01.003. - DOI - PubMed
    1. Lai C.S.W., Franke T.F., and Gan W.B. (2012). Opposite effects of fear conditioning and extinction on dendritic spine remodelling. Nature 483, 87–92. 10.1038/NATURE10792. - DOI - PubMed
    1. Peters A.J., Chen S.X., and Komiyama T. (2014). Emergence of reproducible spatiotemporal activity during motor learning. Nature 510, 263–267. 10.1038/NATURE13235. - DOI - PubMed

Publication types