Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jul 1;118(1):131-139.
doi: 10.1152/jn.00096.2017. Epub 2017 Mar 29.

Development and Long-Term Integration of MGE-lineage Cortical Interneurons in the Heterochronic Environment

Affiliations
Free PMC article

Development and Long-Term Integration of MGE-lineage Cortical Interneurons in the Heterochronic Environment

Phillip Larimer et al. J Neurophysiol. .
Free PMC article

Abstract

Interneuron precursors transplanted into visual cortex induce network plasticity during their heterochronic maturation. Such plasticity can have a significant impact on the function of the animal and is normally present only during a brief critical period in early postnatal development. Elucidating the synaptic and physiological properties of interneuron precursors as they mature is key to understanding how long-term circuit changes are induced by transplants. We studied the development of transplant-derived interneurons and compared it to endogenously developing interneurons (those that are born and develop in the same animal) at parallel developmental time points, using patch-clamp recordings in acute cortical slices. We found that transplant-derived interneurons develop into fast-spiking and non-fast-spiking neurons characteristic of the medial ganglionic eminence (MGE) lineage. Transplant-derived interneurons matured more rapidly than endogenously developing interneurons, as shown by more hyperpolarized membrane potentials, smaller input resistances, and narrower action potentials at a juvenile age. In addition, transplant-derived fast-spiking interneurons have more quickly saturating input-output relationships and lower maximal firing rates in adulthood, indicating a possible divergence in function. Transplant-derived interneurons both form inhibitory synapses onto host excitatory neurons and receive excitatory synapses from host pyramidal cells. Unitary connection properties are similar to those of host interneurons. These transplant-derived interneurons, however, were less densely functionally connected onto host pyramidal cells than were host interneurons and received fewer spontaneous excitatory inputs from host cells. These findings suggest that many physiological characteristics of interneurons are autonomously determined, while some factors impacting their circuit function may be influenced by the environment in which they develop.NEW & NOTEWORTHY Transplanting embryonic interneurons into older brains induces a period of plasticity in the recipient animal. We find that these interneurons develop typical fast-spiking and non-fast-spiking phenotypes by the end of adolescence. However, the input-output characteristics of transplant-derived neurons diverged from endogenously developing interneurons during adulthood, and they showed lower connection rates to local pyramidal cells at all time points. This suggests a unique and ongoing role of transplant-derived interneurons in host circuits, enabling interneuron transplant therapies.

Keywords: fast-spiking neuron; low-threshold-spiking neuron; medial ganglionic eminence.

Figures

Fig. 1.
Fig. 1.
Transplant-derived interneurons develop and maintain the physiological properties characteristic of their lineage. A: whole cell recordings of fluorescently labeled transplant-derived interneurons were obtained at several delays after transplantation (schematic of ages and delays, Aa; example recording, Ab). B: by adulthood [100+ days after transplantation (DAT)], transplant-derived interneurons were either fast spiking (FS) (Ba) or non-FS (Bb) as determined by their firing rate at twice rheobase (Bc; FS cell in dark red, non-FS cell in light red; black outlines indicate responses to twice rheobase). Population data (n = 42 cells from n = 4 mice) demonstrate clear separation of FS and non-FS cells at this age (Bd; gray line indicates twice rheobase stimulus intensity). C and D: this distinction is present as early as 22 DAT (Cc), as it is at intermediate time points (D; dark red dots indicate FS cells, light red indicate non-FS cells, open red circles are too immature to categorize; dashed line indicates threshold used for FS vs. non-FS categorization). E–P: FS and non-FS neurons had different physiological properties throughout development (E–P, dark red dots indicate FS cells, light red indicate non-FS cells, open red circles are too immature to categorize, black outlines indicate example cells in B and C). Dashed line in P marks an interspike interval (ISI) ratio of 1 (i.e., perfectly nonadapting and nonaccelerating firing). To identify parameters that were significantly different between FS and non-FS cells, statistics were computed as follows: for each parameter, a 2-dimensional ANOVA (age and FS/non-FS) was performed, and the resulting set of 9 P values for FS/non-FS was adjusted for multiple comparisons with the Benjamini-Hochberg procedure. For each parameter (since all ANOVAs remained significant), for each of the 4 age groups at which FS and non-FS could be physiologically discriminated, a post hoc Kolmogorov-Smirnov test of FS vs. non-FS was performed. The resulting set of 36 post hoc P values was then multiple-comparisons corrected with the Benjamini-Hochberg procedure. Magnitude of adjusted P values: *P < 0.05, **P < 0.01, ***P < 0.001. Corrected P values: F: P < 0.001, P = 0.12, P < 0.001, P = 0.0053; G: P < 0.001, P = 0.030, P < 0.001, P < 0.001; H: P < 0.001, P = 0.44, P = 0.021, P < 0.001; J: P = 0.0021, P = 0.20, P = 0.040, P = 0.044; K: P < 0.001, P = 0.0035, P < 0.001, P < 0.001; L: P < 0.001, P = 0.015, P = 0.011, P = 0.022; N: P = 0.0037, P = 0.010, P = 0.0025, P < 0.001; O: P < 0.001 for all time points; P: P < 0.001, P = 0.0017, P < 0.001, P < 0.001. WT, wild type; Iinj, injected current; Vm, membrane potential; τ, time constant; AHP, afterhyperpolarization.
Fig. 2.
Fig. 2.
Transplant-derived interneurons develop with a time course similar to endogenously developing interneurons. A: we obtained whole cell recordings from transplant-derived [reds; n (cells, animals) = 46, 5 juvenile; 38, 5 pre-CP; 39, 5 mid-CP; 31, 14 post-CP; 42, 4 adult] and endogenously developing [blues; n (cells, animals) = 52, 4 juvenile; 84, 6 pre-CP; 39, 5 mid-CP; 52, 8 post-CP; 53, 4 adult] interneurons, matched by developmental age. B–J: resting potential (B), input resistance (C), and spike width (F) were significantly lower for transplant-derived interneurons, compared with endogenously developing interneurons, at the earliest developmental time point, while rheobase (H), gain (I), and maximum firing rate (J) were significantly higher, suggesting more rapid maturation. Fast-spiking transplant-derived interneurons (dark red), compared with endogenous fast-spiking interneurons (blue), had lower rheobases (H) and maximum firing rates (J) at the adult time point, while time constant (D), spike threshold (E), and spike width (F) were higher. Fast-spiking neurons also had higher gains (I) and input resistance (C) throughout development. To identify parameters that were significantly different between transplanted and endogenous cells, statistics were computed as follows: for each parameter, separate 2-dimensional ANOVAs (age and endogenous/transplanted) were performed for the fast-spiking and non-fast-spiking populations. The resulting set of 18 P values for endogenous/transplanted was adjusted for multiple comparisons with the Benjamini-Hochberg procedure. For parameters for which the ANOVA remained significant, a post hoc Kolmogorov-Smirnov test of endogenous vs. transplanted was performed for each of the 4 ages at which FS and non-FS could be physiologically discriminated. A Kolmogorov-Smirnov test of endogenous vs. transplanted was also performed for the juvenile time point. The resulting set of 41 P values was then multiple-comparisons corrected with the Benjamini-Hochberg procedure. Adjusted P values are indicated above each age/parameter combination (green = P < 0.05, darker green = lower P value). “NA” indicates parameters for which post hoc tests were not performed because the initial ANOVA was not significant. Error bars represent SE.
Fig. 3.
Fig. 3.
Transplant-derived interneurons synapse onto host pyramidal neurons with unitary connections that are similar to those of endogenously developing interneurons. A: we obtained paired whole cell recordings between pyramidal cells and either transplant-derived interneurons (n = 16 animals at each time point) or endogenously developing interneurons (n = 8 animals at each time point). B–E: paired recordings from host pyramidal cells (gray) and fluorescently identified endogenous (blue, B and D; interneurons are both FS) or transplant-derived (red, C and E; interneurons are FS and non-FS, respectively) interneurons at equivalent times after interneuronogenesis reveal monosynaptic inhibitory postsynaptic potentials (IPSPs) from interneurons onto host pyramidal neurons. F–H: IPSPs are small (F), depressing (G; 50-ms presynaptic interspike interval), and of brief latency (H). For each age, the difference between endogenous vs. transplanted was tested with a Kolmogorov-Smirnov test [amplitude: P = 0.38, 0.99; PPR: P = 0.93, 0.79; latency: P = 0.38, 0.14 (for mid-CP and post-CP, respectively)]. I: connection probabilities from paired recordings show that transplant-derived neurons are less likely to contact host pyramidal cells than are endogenously developing interneurons (*P = 0.031 by 2-tailed Cochran-Mantel-Haenszel exact test). Numbers within bars: no. of observed connections (top) vs. no. of connections tested (bottom).
Fig. 4.
Fig. 4.
Transplant-derived neurons receive excitatory postsynaptic potentials (EPSPs) from host pyramidal cells that are similar to EPSPs onto host interneurons and are present into adulthood. A–D: paired recordings from host pyramidal cells (gray, A–D) and fluorescently identified endogenous (blue, A and C; recordings from 8 animals at each time point) or transplant-derived (red, B and D; recordings from 16 animals at each time point) interneurons at equivalent times after interneuronogenesis (A and B, mid-CP; C and D, post-CP) reveal monosynaptic EPSPs from host pyramidal neurons onto interneurons. E–G: quantification of amplitude (E), paired-pulse ratio (F), and synaptic delay (G). For each age, the difference between endogenous vs. transplanted was tested with a Kolmogorov-Smirnov test [amplitude: P = 0.42, 0.34; PPR: P = 0.59, 0.66; latency: P = 0.68, 0.22 (for mid-CP and post-CP, respectively)]. H: connection probability was not different (P = 0.23 by 2-tailed Cochran-Mantel-Haenszel exact test). Numbers within bars: no. of observed connections (top) vs. no. of connections tested (bottom). I–K: the rate of spontaneous excitatory inputs [spontaneous excitatory postsynaptic currents (EPSCs)] onto transplant-derived interneurons [n (cells, animals) = 45, 5 juvenile; 36, 5 pre-CP; 39, 5 mid-CP; 17, 8 post-CP; 42, 4 adult) was lower than that of endogenously developing interneurons of the same age [I; n (cells, animals) = 52, 4 juvenile; 83, 6 pre-CP; 36, 5 mid-CP; 48, 8 post-CP; 53, 4 adult] but not their amplitude (J) or rise time (K). To identify parameters that were significantly different between transplanted and endogenous cells, statistics were computed as follows: for each parameter, separate 2-dimensional ANOVAs (age and endogenous/transplanted) were performed for the fast-spiking and non-fast-spiking populations. The resulting set of 6 P values for endogenous/transplanted was adjusted for multiple comparisons with the Benjamini-Hochberg procedure. For parameters for which the ANOVA remained significant, a post hoc Kolmogorov-Smirnov test of endogenous vs. transplanted was performed for each of the 4 ages at which FS and non-FS could be physiologically discriminated. A Kolmogorov-Smirnov test of endogenous vs. transplanted was also performed for the juvenile time point. The resulting set of 11 P values was then multiple-comparisons corrected with the Benjamini-Hochberg procedure. The adjusted P values are indicated above each age/parameter combination (green = P < 0.05, darker green = lower P value). “NA” indicates parameters for which post hoc tests were not performed because the initial ANOVA was not significant. Error bars represent SE.

Similar articles

See all similar articles

Cited by 5 articles

Publication types

LinkOut - more resources

Feedback