Differential Glutamatergic Inputs to Semilunar Granule Cells and Granule Cells Underscore Dentate Gyrus Projection Neuron Diversity

Differential Glutamatergic Inputs to Semilunar Granule Cells and Granule Cells Underscore Dentate Gyrus Projection Neuron Diversity | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Differential Glutamatergic Inputs to Semilunar Granule Cells and Granule Cells Underscore Dentate Gyrus Projection Neuron Diversity Laura Dovek , Anh-Tho Nguyen , Emmanuel Green , View ORCID Profile Vijayalakshmi Santhakumar doi: https://doi.org/10.1101/2025.03.14.643192 Laura Dovek 1 Department of Molecular, Cell and Systems Biology, University of California Riverside , Riverside, California 92521 Find this author on Google Scholar Find this author on PubMed Search for this author on this site Anh-Tho Nguyen 1 Department of Molecular, Cell and Systems Biology, University of California Riverside , Riverside, California 92521 Find this author on Google Scholar Find this author on PubMed Search for this author on this site Emmanuel Green 1 Department of Molecular, Cell and Systems Biology, University of California Riverside , Riverside, California 92521 Find this author on Google Scholar Find this author on PubMed Search for this author on this site Vijayalakshmi Santhakumar 1 Department of Molecular, Cell and Systems Biology, University of California Riverside , Riverside, California 92521 Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Vijayalakshmi Santhakumar For correspondence: vijayas{at}ucr.edu Abstract Full Text Info/History Metrics Supplementary material Preview PDF Abstract Semilunar Granule Cells (SGCs) are sparse dentate gyrus projection neurons whose role in the dentate circuit, including pathway specific inputs, remains unknown. We report that SGCs receive more frequent spontaneous excitatory synaptic inputs than granule cells (GCs). Dual GC-SGC recordings identified that SGCs receive stronger medial entorhinal cortex and associational synaptic drive but lack short-term facilitation of lateral entorhinal cortex inputs observed in GCs. SGCs dendritic spine density in proximal and middle dendrites was greater than in GCs. However, the strength of commissural inputs and dendritic input integration, examined in passive morphometric simulations, were not different between cell types. Activity dependent labeling identified an overrepresentation of SGCs among neuronal ensembles in both mice trained in a spatial memory task and task naïve controls. The divergence of modality specific inputs to SGCs and GCs can enable parallel processing of information streams and expand the computational capacity of the dentate gyrus. Introduction The hippocampus plays a critical role in the formation and consolidation of new memories. In particular, the dentate gyrus (DG) is proposed to regulate the flow and integration of multisensory input streams to the hippocampus and promote the formation of distinct memory traces. Studies in the hippocampus and cortex have revealed that seemingly homogeneous neurons can exhibit subtle diversity in structure, intrinsic physiology, and connectivity which can enhance the information processing in the circuit ( Slomianka et al., 2011 ; Soltesz and Losonczy, 2018 ). Recent studies have identified that a subtype of dentate projection neuron, semilunar granule cells (SGCs) have several features that distinguish them from the classical granule cells (GCs). SGCs were first identified by their location above the granule cell layer (GCL), their wide dendritic spread and characteristic “semilunar” somatic shape ( Williams et al., 2007 ; Gupta et al., 2012 ; Afrasiabi et al., 2022 ). Compared to GCs, SGCs have lower input resistance and less adapting firing pattern ( Williams et al., 2007 ; Afrasiabi et al., 2022 ; Dovek et al., 2024 ). SGCs receive more frequent inhibitory postsynaptic currents (IPSCs) and have higher extrasynaptic GABA currents ( Gupta et al., 2012 ; Gupta et al., 2022 ). Similar to CA1 pyramidal cell subtypes, which are generated in different waves of neurogenesis ( Bayer, 1980 ), SGCs are preferentially generated during late embryonic generation of granule cell development ( Save et al., 2019 ). Although SGCs are a relatively sparse subtype ( Save et al., 2019 ), they are disproportionately labeled as part of activity-dependent behavioral engrams, suggesting a preferential role in dentate processing ( Erwin et al., 2020 ; Dovek et al., 2024 ). However, whether SGCs and GCs differ in their glutamatergic synaptic inputs is not fully known. Afferent drive to the DG segregates into layered and functionally distinct input streams that are integrated in GC and SGC dendrites to evoke firing. Massive cortical inputs from the entorhinal cortex (EC) impinge on the DG through the perforant path (PP) afferent fibers. The PP segregates into two stratified input streams with distinct information content: the medial perforant path arising from the medial entorhinal cortex (MEC) projects to GC and SGC dendrites in the middle molecular layer (MML), and the lateral perforant path arising from the lateral entorhinal cortex (LEC) targets distal dendrites in the outer molecular layer (OML). While MEC is involved in processing spatial (where/context) information, the LEC carries information about the object (what/content) ( Knierim et al., 2006 ; Knierim et al., 2014 ; Fernández-Ruiz et al., 2021 ). Apart from PP inputs, axons of glutamatergic mossy cells (MCs) in the DG hilus, extend robust commissural and associational projections along the septotemporal axis of the DG terminating on the contralateral and ipsilateral GC dendrites, respectively ( Buckmaster et al., 1996 ; Houser et al., 2020 ; Botterill et al., 2021 ). Ventral mossy cells project dense inputs to the proximal dendrites of GCs in the inner molecular layer (IML). However, recent studies have identified that dorsal mossy cell terminals show a more diffuse distribution overlapping with MEC inputs in the MML ( Houser et al., 2020 ; Botterill et al., 2021 ). In addition to their differing projections, ventral mossy cells are activated by environmental novelty, such as context discrimination, and preferentially project to dorsal GCs ( Fredes et al., 2021 ). Yet, the functional characteristics of dorsal and ventral mossy cells projections to dentate cell types has not been evaluated. Previous studies have found that, while the amplitude of perforant path driven excitatory currents are not different between GCs and SGC, the stimulation of the hilus elicits larger currents in SGCs ( Williams et al., 2007 ). Whether these individual excitatory input streams differentially influence SGCs and GCs is currently unknown. Neuronal activation is determined both by the strength of the dendritic inputs and by how the inputs are integrated in the dendrites. Integration and attenuation of inputs is shaped by cell morphology and dendritic active and passive membrane properties. In this regard, GCs have been found to demonstrate unique dendritic input propagation with limited input attenuation in the distal and mid dendritic segments and robust signal attenuation at the proximal dendrite. These dendritic features result in strong dendritic input integration in GCs, not observed in other excitatory neurons ( Krueppel et al., 2011 ). Computational studies have identified that morphological features of GCs contribute to their unique dendritic integration and attenuation characteristics. Given the morphological differences between GCs and SGCs, it is important to determine whether neuronal structure impacts input attenuation and input integration in the two cell types. We examined the spontaneous glutamatergic drive and layer specific inputs to GCs and SGCs. We find that SGCs receive significantly more spontaneous excitatory synaptic inputs than GCs. Optically driven recordings of excitatory synaptic inputs to GCs - SGCs pairs identified cell-type and pathway specific differences in strength and short-term plasticity of entorhinal cortex inputs to SGCs and GCs. Simulation in morphologically constrained detailed multicompartmental models of GCs and SGCs showed that dendritic input integration was not different between the cell types. Consistent with receiving higher spontaneous EPSCs and stronger MPP inputs, activity dependent labeling of neurons revealed greater proportional recruitment of SGCs to neuronal ensembles both in task naïve control mice in their home cage and following a dentate-dependent spatial memory task. Our results extend the known difference between GCs and SGCs to their circuit connectivity and support the proposal that SGCs and GCs form a parallel interacting dentate microcircuits, enhancing information processing in the dentate gyrus. RESULTS Semilunar granule cells receive more frequent spontaneous glutamatergic inputs Recent studies have identified differences in intrinsic physiology and inhibitory synaptic inputs between GCs and SGCs ( Williams et al., 2007 ; Gupta et al., 2012 ; Gupta et al., 2020 ; Afrasiabi et al., 2022 ; Dovek et al., 2024 ). However, little is known about the steady state excitatory drive to SGCs. To address this knowledge gap, we recorded spontaneous excitatory post synaptic currents (sEPSCs) in GCs and SGCs. SGCs were identified by the presence of multiple primary dendrites and greater dendritic angle, as well as soma location and larger soma aspect ratio compared to GCs, as described previously ( Gupta et al., 2020 ; Afrasiabi et al., 2022 ) ( Fig 1A-C ). A summary plot of the number of primary dendrites and dendritic angle in the recorded neurons revealed the emergence of two groups, aligning with experimenter defined cell categorization as GCs (black) or SGCs (blue) ( Fig 1C ). These data confirm the ability to reliably distinguish GCs and SGCs based on morphological criteria. Any cell not clearly categorized into either group was excluded from further analysis. Voltage clamp recordings from a holding potential of -70mV identified that SGCs received more frequent sEPSCs than GCs ( Fig 1D,E , frequency in Hz GCs: 4.78 ± 0.25 N=14/6 mice, SGCs: 7.45 ± 0.38 N=13/7 mice, p<0.0001, Kolmogorov-Smirnov Test). However, sEPSC amplitude, charge transfer, 10-90% rise time and decay were not different between cell types. ( Fig 1F,G , Supplemental Fig 1A-C,). Taken together with our prior finding that SGCs receive more inhibitory synaptic inputs than GCs ( Gupta et al., 2012 ; Afrasiabi et al., 2022 ), these results indicate that SGCs receive more spontaneous synaptic inputs than GCs. Download figure Open in new tab Figure 1: SGCs receive more frequent spontaneous excitatory inputs than GCs. A-B) Representative images of a biocytin filled GC with a narrow dendritic span and a single primary dendrite (A) and a SGC with wide dendritic span and multiple primary dendrites (B). Maximum intensity projections of confocal image stacks are presented as gray-scale, inverted images to visualize hilar axon collaterals. C) Summary plot of dendritic angle and primary dendrites of cells classified as SGC (blue) and GC (black) by experimenter. D) Representative current traces illustrate spontaneous EPSCs in a GC (top) and SGC (bottom). E) Cumulative probability plot of sEPSC instantaneous frequency in GCs (black) and SGCs (blue). F) Example of average sEPSC in representative GC (black) and SGC (blue) overlaid on individual traces in gray. G) Summary plot of average sEPSC amplitude. * indicates p<0.05 by Kolmogorov-Smirnov test, n=13-14 cells/group. SGCs receive stronger medial entorhinal cortex inputs than in GCs The perforant path is the primary source of cortical inputs to GCs and SGCs. Consistent with previous findings in rats ( Williams et al., 2007 )., the EPSC amplitude evoked by mass stimulation of the perforant path was not different between GCs and SGCs (Supplemental Fig 2 A-C, SGC n=8 GC n=9 p>0.05). Because the perforant path includes two distinct input streams, we examined the pathway specific inputs to GCs and SGCs by selectively expressing the excitatory opsin, channelrhodopsin (ChR2), in the lateral or medial perforant path projections in slices from mice injected with AAV5-hSyn-hChR2(H134R)-EYFP into the lateral or medial entorhinal cortex. Dual recordings from SGC-GC pairs, with somata located within 200 µm of each other and at similar depths, were conducted to limit potential confounds due to differences in ChR2 expression between animals and slices ( Fig 2 A,F). Download figure Open in new tab Figure 2: SGCs receive stronger MEC inputs compared to GCs. A) Representative image of biocytin filled SGC and GC (red). Lateral perforant path projections are labeled with ChR2-eYFP (green). B) Representative traces of EPSCs evoked by a single 10ms light pulse in a GC (top) and a SGC (bottom). Individual traces are in gray with average trace overlaid in black (GC) or blue (SGC). C) Summary plot of eEPSC peak amplitude in SGC-GC pairs (paired t-test n=9 SGC-GC pairs) D) Representative traces of EPSCs evoked by a train of 10 light pulses at10 Hz in GC (top) and SGC (bottom). Dotted line represents the peak amplitude of the first EPSC. E) Summary plot of the ratio of amplitude of n th evoked EPSC compared to the first EPSC evoked during a 10 pulse train. * indicates p<0.05, for main factor cell type by TW-ANOVA. F) Representative image of biocytin filled SGC and GC (red) following AAV injection to MEC. MEC projections are labeled with ChR2-eYFP(green). G) Representative traces of EPSCs evoked by a single 10ms light pulse in GC (top) and SGC (bottom). H) Summary plot of eEPSC peak amplitude in SGC-GC pair. I) Representative EPSC traces in a GC (top) and a SGC (bottom) evoked by a train of 10 stimuli, 10Hz light pulses to activate MEC projections. J) Summary plot of the ratio of the amplitude of n th evoked EPSCs compared to the first EPSC evoked during a 10 pulse train. * Indicates p<0.05 by paired t-test n=8 SGC-GC pairs. As illustrated by an example image of a biocytin filled GC-SGC pair, AAV injection in the LEC reliably labeled perforant path fibers in the OML with ChR2-eYFP ( Fig 2A , Supplemental Fig 2D). Blue light activation of LEC inputs in the OML (0.9mW, λ=470 nm, 10ms pulse) evoked low amplitude EPSCs with no systematic difference between cell types ( Fig 2B , C; LPP evoked EPSC amplitude p=0.38, n=7 SGC-GC pairs). However, the short-term dynamics of the responses to LEC activation at 10 Hz (10 pulses) differed between cell types. While LEC evoked EPSCs in GCs showed multi-pulse facilitation, EPSCs in SGCs underwent synaptic depression during the same stimulus train ( Fig 2D , E). In control experiments, blue light stimulation failed to elicit synaptic responses when mice were injected with control virus (AAV5-hSyn-GFP) lacking ChR2 (Supplemental Fig. 2E-F). Together these data show that while the LPP input strengths are not different between GCs and SGC, there is a cell-type specific difference in short-term plasticity which can impact dendritic processing of LEC inputs. Next, we examined GC-SGC pairs in mice injected with AAV in the MEC to express ChR2-eYFP in medial perforant path fibers in the MML ( Fig 2F , Supplemental Fig 2G). In contrast to LEC inputs, optical activation of MEC projections consistently evoked higher amplitude EPSCs in SGCs compared to neighboring GCs in the same slice ( Fig 2G-H , p=0.031, n=7 SGC-GC pairs). However, both GCs and SGCs showed similar multi-pulse synaptic depression in responses to optical activation of MPP inputs at 10 Hz. ( Fig 2I,J ). Taken together, the stronger MPP synaptic inputs and multi-pulse depression of both LPP and MPP synapses observed in SGCs are well suited to recruit SGCs during behaviorally relevant MEC activity. Cell-type specific differences in associational but not commissural inputs between SGCs and GCs Apart from the perforant path, mossy cells are a major source of glutamatergic commissural and associational inputs to the DG ( Buckmaster et al., 1996 ). Local axon collaterals of mossy cells provide sparse innervation of hilar neurons and local GCs ( Botterill et al., 2021 ). In contrast, long-range mossy cell axons form associational and commissural projections that innervate ipsilateral and contralateral GCs, respectively, along the entire septotemporal axis of the DG ( Frotscher et al., 1991 ; Buckmaster et al., 1996 ). Consistent with the previous reports in rats ( Williams et al., 2007 ), our slice recordings in mice revealed that the amplitude of hilar stimulation evoked EPSCs is greater in SGCs than in GCs ( Fig 3A-C ) indicating stronger associational projections to SGCs than GCs. Download figure Open in new tab Figure 3: SGCs receive stronger associational but not commissural synaptic inputs than GCs. A) Schematic depicting electrode placement for hilar stimulation. B) Representative EPSCs in a GC (left) and SGC (right) in response to electrical stimulation in the hilus. Individual traces are in gray, with average trace overlaid in black. C) Summary plot of hilar stimulus-evoked EPSC peak amplitudes at increasing intensities n=5-6 cells/group. * Indicates p<0.05 by One-Way ANOVA. D) Representative image of biocytin filled SGC and GC (red) with AAV driven ChR2-eYFP labeling of commissural projections from the contralateral dorsal hilus (green). E) Representative EPSC traces in a simultaneously recorded GC (top)-SGC (bottom) pair evoked by activation of commissural projections from the dorsal hilus by a single 10ms light pulse. Individual traces are in gray, with average trace overlaid in black (GC) or blue (SGC). F) Summary plot of EPSC peak amplitude in SGC-GC dual recordings. p>0.05 paired t-test, n=11 SGC-GC pairs. G) Representative traces of EPSCs evoked by a 10 Hz train of 10ms light pulses, in GC (top) and SGC (bottom). H) Summary plot of the ratio of the amplitude of evoked EPSC compared to the first EPSC evoked in a 10 pulse train. I) Representative image of biocytin filled SGC and GC (red) with AAV driven ChR2-eYFP labeling of commissural projections from the ventral hilus (green). J) Representative EPSC traces evoked by a single 10ms light pulse activation of commissural projections from ventral hilus in GC (top) and SGC (bottom). Individual traces are in gray with average trace overlaid in black (GC) or blue (SGC) in n=10 SGC-GC pairs. K) Summary plot of eEPSC peak amplitude in SGC-GC dual recordings. L) Representative traces of EPSCs evoked by a 10 Hz train of 10ms light pulses in GC (top) and SGC (bottom). M) Summary plot of the ratio of amplitude of evoked EPSCs compared to the first EPSC evoked in a 10 pulse train. * Indicates p<0.05 by paired t-test n=8 SGC-GC pairs. It has been previously demonstrated that the spatial distribution of mossy cell axon collaterals differs depending on the location of their somata along the dorsoventral axis of the DG ( Houser et al., 2020 ; Botterill et al., 2021 ). To how this differential mossy cell axonal distribution activates GCs and SGCs, we independently evaluated ventral and dorsal commissural inputs to the cell types. We injected AAV5-CAMKIIa-hChR2(H134R)-eYFP unilaterally into the dorsal or ventral hilus to selectively label dorsal and ventral glutamatergic commissural inputs to the contralateral DG (Supplemental Fig 3A, B). AAV injection in the dorsal hilus consistently labeled diffuse commissural projections extending through the inner and middle molecular layers in the contralateral hilus ( Fig 3D ). Optical activation of the dorsal commissural inputs evoked EPSCs in both cell types ( Fig 3E ). The amplitude ( Fig 3F ) and multi-pulse depression in response to activation of dorsal commissural inputs at 10Hz (10 stimuli each with 10 ms duration) were not different between cell types ( Fig 3G, H ). As reported previously ( Houser et al., 2020 ; Botterill et al., 2021 ), AAV injection in the ventral hilus (Supplemental Fig 3B) labeled a distinct band of fibers restricted to the inner molecular layer in the contralateral hilus ( Fig. 3I ). As with dorsal commissural projections, recordings from GC-SGC pairs revealed no differences in EPSC amplitude ( Fig 3J , K) or multi-pulse depression ( Fig 3L , M) between GCs and SGCs. In control experiments in which mice were injected with AAV5-CAMKIIa-GFP lacking ChR2 in the ventral hilus, optical activation failed to elicit synaptic responses in the recorded GCs (Supplemental Fig. 3C, D). Collectively, these data demonstrate that while SGCs receive stronger associational inputs than GCs, the commissural inputs from the dorsal and ventral hilus are not different between cell types. Dendritic spine density is higher in SGCs than in GCs In light of the higher sEPSC frequency and stronger MEC inputs, we examined the density of dendritic spines in biocytin-filled SGCs and GCs obtained during physiological recordings. Images were captured from 3 dendritic segments: proximal, middle, and distal to the soma (illustrated in Fig 4A-B ). Remarkably, SGCs but not GCs had spines on both somata and on putative axons ( Fig 4C ) which are well positioned to receive associational inputs in the inner molecular layer. Dendritic spine reconstructions revealed higher overall spine density in SGCs than in GCs ( Fig 4D ; Spines / µm: SGC 1.12±0.06, GC 0.81±0.08, p=0.0134 by nested t-test. n=6 cells/group). Evaluation of spine density relative to distance from soma, identified that the spine density in the middle and proximal dendritic segments of SGCs was significantly higher than in GCs ( Fig 4A-B, E ). Note that because of their wide dendritic arbor and somatic location in the inner molecular layer, a majority of SGC proximal and middle dendrites are in the middle molecular layer which receives MPP inputs. However, spine density in the distal dendritic segments in the outer molecular layer innervated by the LPP, was not different between the cell types. Download figure Open in new tab Figure 4: SGCs have higher dendritic spine density than GCs. A-B) Representative confocal image of a GC (A) and a SGC (B) with insets illustrating reconstructed spines in proximal (yellow), middle (purple) and distal (green) dendritic segments. C) Representative images of 3 SGC somata demonstrating the presence of somatic spines (white arrows) as well as axonal spines (yellow arrows). D) Summary plot compares spine density between cell types. * Indicates p<0.05 by Mann Whitney n=18 segments/6cells/group. E) Summary plots of layer-specific differences in spine densities between cell types. F) Summary plots of densities of thin, stubby, mushroom and filapodial spines compared between cell types. * Indicates p<0.05, ** Indicates p<0.01 by Two Way-ANOVA with Šídák’s multiple comparisons post hoc test in n=6 cells/group. Spines can be categorized based on their shape and spine morphology is associated with distinct functional characteristics. We examined thin, stubby, mushroom, and filapodial spine types, classified using automated algorithms in Neurolucida 360, followed by validation by the experimenter, to determine if SGCs preferentially express an excess of a specific dendritic spine subtype. Thin spines, which can be transient and are referred to as “learning spines” due to their proposed involvement in plasticity ( Popov et al., 2004 ; Bourne and Harris, 2007 ), were more abundant in SGCs compared to GCs ( Fig 4F ). Additionally, the SGC somatic spines appear to be a mix of filopodial and thin spines. Densities of dendritic stubby, mushroom. and filapodial spines were not different between cell types. In addition to differences in dendritic arbors and spine density, our cell fills revealed that SGC axons often projected extensively throughout the IML and GCL prior to entering the hilus, a unique feature of SGCs previously reported in rats (Supplemental Fig 4) ( Williams et al., 2007 ). Collectively, these data identify that SGCs with somatic spines and increased proximal and middle dendritic spines are well positioned to receive more associational and medial perforant path inputs. Cell morphology does not impose differences in dendritic EPSP integration between GCs and SGCs Given the layer-specific inputs and the potential role for passive dendritic structures in regulating synaptic signal processing, we implemented morphologically constrained multicompartmental models of GCs and SGCs to evaluate whether the neuronal structure could contribute to differences dendritic attenuation and integration of synaptic inputs in GCs and SGCs. Accurate 3D reconstructions of 5 biocytin filled GCs and SGCs, generated using Neurolucida 360, were exported into “Trees” software ( Cuntz et al., 2011 ) to segment dendrites into functionally distinct thirds representing the inner, middle and outer molecular layer synaptic input zones. Segmented GC/SGC morphologies were imported into NEURON ( Hines and Carnevale, 1997 ) to implement active and passive conductances to enable functional simulations ( Santhakumar et al., 2005 ). Channel densities including leak conductance were maintained identical between all neurons to isolate the impact of neuronal morphology. Consistent with the difference in soma shape between GC and SGC, the somatic compartment in the SGC was modeled to be larger than in GCs. The resting membrane potential (RMP) of model cells was set at -70 mV and was not different between model and biological cells (RMP in mV, model GC: 70.51±0.14, n=4; biological GC: 78.25±3.86, n=4; model SGC: 70.25±0.05, n=4; biological SGC: 72.76±3.32, n=5, p= 0.351 F (1, 13) = 0.94). Similar to biological cells ( Williams et al., 2007 ; Gupta et al., 2012 ; Afrasiabi et al., 2022 ; Dovek et al., 2024 ), the input resistance (R in ) of model SGCs was lower than in the model GCs (input resistance in MΩ, model GC 179.4±18.97; model SGC: 99.95±18.77, n=4; p<0.05). Identical biexponential synaptic conductance simulating AMPA synapses ( Golding et al., 2005 ), were located in dendritic segments corresponding to the three major input layers (IML, MML, and OML). The local (dendritic) and somatic voltage changes (EPSPs) in response to activation of the synaptic conductance were measured to evaluate the effect of morphology on dendritic attenuation of EPSP amplitude between cell types ( Fig. 5A ). Comparison of the local dendritic voltage revealed that while local dendritic amplitude of synapses in the proximal dendrite was not different between GC and SGC, the local EPSP evoked by an identical conductance was significantly higher in GC than in SGC in both middle and distal dendritic locations ( Fig 5B ). Moreover, the proximal to distal boosting of dendritic amplitude, while present, was considerably attenuated in SGCs ( Fig 5B ). Interestingly, within each cell type, the somatic EPSP amplitude was not different regardless of the proximal-distal location of the synapse ( Fig. 5C ), consistent with the lack of location dependence in GC input integration ( Krueppel et al., 2011 ). However, the peak amplitude of somatic EPSPs in response to an identical dendritic conductance was significantly higher in GCs than in SGCs ( Fig 5C ). Somatic EPSP amplitude measured as a percent of the corresponding local dendritic EPSP, was lower when synapses were positioned farther away from the soma consistent with increased attenuation of distal synapses. Despite the significant effect of dendritic distance on input attenuation, there were no cell type specific differences in the extent of amplitude attenuation ( Fig 5D ). Download figure Open in new tab Figure 5: Morphometric computational models reveal greater dendritic boosting of EPSP in GCs than in SGCs. A) Representative reconstruction of a GC (top) and a SGC (bottom) and membrane voltage traces show EPSPs in response to activation of identical AMPA conductance at various dendritic locations. Traces illustrate EPSP amplitude at the soma (black) and at the local input site in the IML (green), MML (blue), and OML (purple). B-C) Amplitude of EPSPs at the site of input on the dendritic segment (B) and at the soma (C). **indicates p<0.01 **** indicates p<0.0001 by TW-ANOVA with Šídák’s multiple comparisons post hoc test in n= 3 trials/4 cells. D) Summary plot of attenuation comparing amplitude at the site of the stimulus to the amplitude recorded ad the soma. E) Representative reconstructions of GC left and SGC right illustrates location of 5 input synapses in the middle molecular layer on either different branches selected at random (top) or on the same dendritic branch (bottom). F-G) Summary plot of the summated EPSP amplitude in GCs (F) and SGCs (G) when the stimuli were on either the same branch or different branches. N=10 trials/ cell, 4 cells/group. Since we identified stronger MPP inputs to SGC, we focused on the middle dendritic segment and examined the summation of inputs to the same or different dendritic branches ( Fig 5E ). We distributed five identical AMPA synaptic conductances either within a single middle dendritic segment on the same branch (5 trials/cell) or randomly in middle dendritic segments across dendritic branches (5 trials/cell). While the input summation was sublinear in both cell types, there was no difference in the summation of synaptic inputs within and between dendritic branches in either cell type ( Fig 5F-G ). Overall, the simulations indicate that dendritic structure does not differentially impact synaptic input attenuation or summation in GCs and SGCs. Preferential recruitment of SGCs to DG neuronal ensembles is independent of behavior task Since we find that SGCs have a greater basal excitatory drive as well as a selectively robust MPP input compared to GCs, we adopted TRAP2::tdT transgenic mice to induce c-Fos driven tdT (reporter) labeling of DG ensembles ( DeNardo et al., 2019 ) in order to evaluate competing hypotheses concerning SGC recruitment to active neuronal ensembles. Specifically, we sought to test whether the more frequent spontaneous glutamatergic drive and sustained firing characteristics rendered SGCs more active regardless of task, or if SGCs were selectively activated by the MPP contextual input stream. Since SGCs represent approximately 5% of the GC population ( Save et al., 2019 ), we reasoned that a greater than 5% reporter labeling in SGCs in control mice housed in the home cage with a reduction in proportional activation during a contextual task would suggest that SGCs have elevated activity levels unrelated to behavioral tasks. Alternatively, a low 5% basal SGC labeling in task naïve mice with labeling of a greater proportion of SGCs as part of active neuronal ensembles in mice trained in a contextual spatial learning task would suggest selective SGC activation by contextual information. Interestingly, while prior studies have demonstrated greater than expected activity-dependent labeling of SGCs in various hippocampal and dentate behavioral tasks, their relative recruitment did not appear to depend on the nature of the task ( Erwin et al., 2020 ; Dovek et al., 2024 ). We adopted task naïve controls and littermate mice trained in a DG-dependent Barnes maze task to evaluate the recruitment of SGCs to active neuronal ensembles. Experimental mice were trained for 6 days on the Barnes maze to use spatial cues to find an escape followed by a probe trial and 2 days of reacquisition trials (Re-aq) ( Fig 6A ). Task-naïve littermates were brought to the testing room but were never exposed to the Barnes maze table. Evaluation of primary latency, primary errors, total latency, and total errors for mice to locate the escape box confirmed the progressive improvements in task performance from acquisition days 1 through 6 (Supplemental Fig 5A-D). BUNS analysis ( Illouz et al., 2016 ) to assess the use of spatial search strategy, revealed that the mice transitioned from using a random or serial search strategy to a spatial strategy as they progressed through acquisition days. A concatenated version of their search strategy combining all strategies that involved some aspect of spatial awareness into a “spatial approach” in contrast to the serial and random search approaches further confirmed the maximum use of spatial strategy by day 6 ( Fig 6B , Supplemental Fig 5E). Additionally, cognitive scores based on the strategy (developed by Illouz et al ., 2016 ) demonstrated progressive improvement from days 1 through 6 of acquisition ( Fig 6C , p<0.00001 by Kruskal Wallis ANOVA with Dunn’s multiple comparisons n= 14 mice). Having established that test animals reliably adopt a spatial search strategy by day 6, we injected 4-hydroxytamoxifen (4-OHT) 10 minutes before day 6 of the acquisition trial to induce tdT expression in neurons expressing the immediate early gene c-Fos during behavior testing. Littermate controls in their home cage were also injected with 4-OHT and returned to the home cage. To minimize non-specific activation, test mice and their littermate controls were housed in the test room for 5 hours prior to and following injection and testing, before returning to the vivarium. A probe trial on day 7, during which the escape was removed, showed that mice with acquisition training spent significantly more time in the quadrant in which the escape was previously located compared to untrained littermate control animals ( Fig 6D ), confirming the use of spatial search strategy in trained animals. Download figure Open in new tab Figure 6: Semilunar granule cells are reliably labeled during cFos-driven tagging of DG neurons. A) Schematic of experimental timeline for test mice. B) Summary of search strategies used by mice in the Barnes maze task. C) Summary of cognitive scores, derived from search strategy used during task acquisition and probe trials. D) Summary plot of overall time spent in escape quartile during the probe trial. *** indicates p<0.001 by Mann-Whitney. E-F) Representative epifluorescence image of sections from mice one week after induction of tdT labeling (red) followed by c-Fos immunostaining (green) after a reacquisition trial. Control mice were retained in a home cage and not exposed to behavioral testing (E). Test mice were treated with tamoxifen on day 6 of acquisition and euthanized after reacquisition for c-Fos labeling (F). White arrows identify cells co-labeled for tdT and c-Fos. G) Quantification of tdT labeled cells in sections from control and Barnes Maze trained mice. H) Quantification proportion of cells labeled with tdT during acquisition that were co-labeled with c-FOS (green) following reacquisition (re-aq) I) Proportion of granule cells tagged with tdT that are morphologically classified as SGCs (H). * indicates p<0.05 by nested t-test. Schematics were generated using BioRender under license. Having established that, unlike task naïve controls, mice trained in the Barnes maze reliably adopt a spatial strategy, we quantified the number of neurons labeled with tdT in sections from animals labeled on day 6 of acquisition and task naïve littermate control animals. While there was considerable variability in the number of tdT labeled neurons between animals, there was a trend towards more labeled neurons in the DG of trained mice than in littermate controls ( Fig 6E-G ; Test: 11.62 ±1.98 Control: 5.84±0.65, p=0.056 Mann-Whitney, n=5 animals/group). To verify that the tdT labeled cells were recruited during memory recall, animals underwent re-acquisition trials and were sacrificed 90 minutes later for immunostaining for c-Fos. The reacquisition trials identified that the ability of mice to locate the escape box measured by latency to locate the box, the number of errors (Supplementary Fig. 5A-D). Moreover, the search strategies used to locate the escape box during re-acquisition trials were not different from acquisition day 6 ( Fig. 6b , Supplementary Fig. 5A-D). Consistently, compared to controls, trained mice showed significantly greater colocalization of tdT with c-Fos indicating reactivation of the same cells during memory recall ( Fig 6H ; Co-Labeling: Test 0.17 ± 0.03 Control 0.03 ± 0.013, p<0.0001 Mann-Whitney test, n=50 sections from 5mice/group). These results confirm that, unlike task naïve controls, a significant proportion of DG neurons labeled with tdT in the mice trained on the Barnes maze are activated and reactivated in response to spatial navigation. Having confirmed task-dependent tdT labeling in mice trained in the Barnes maze, we adopted morphologically based classification of tdT labeled neurons to quantify the number of labeled GCs and SGCs. SGCs were identified based on the presence of multiple primary dendrites, wide dendritic arbor, and greater soma width compared to height and location in or near the inner molecular layer ( Afrasiabi et al., 2022 ; Dovek et al., 2024 ). Despite comprising only an estimated 5% of the GC population ( Save et al., 2019 ), approximately 25% of the tdT labeled cells in task naïve mice were SGCs indicating higher basal activity levels in SGCs compared to GCs. Interestingly, the proportion of SGCs labeled as part of active neuronal ensembles was not different between naïve controls and trained littermate mice ( Fig 6I ; %SGC: Control 28.65 ± 3.48, Trained: 23.47 ± 3.06; p= 0.25 Mann-Whitney test, n= 50 sections from 5 mice/ group). These findings suggest that higher basal glutamatergic drive and intrinsic excitability may support disproportionate labeling of SGCs as part of active neuronal ensembles in control mice. Concurrently, the higher proportional SGC labeling is maintained during spatial behavioral tasks suggesting SGC recruitment during spatial navigation. Discussion There is considerable interest in understanding how the dentate gyrus functions due to the key role for this circuit as a critical node in entorhinal-hippocampal information transfer and in consolidating inputs from several sensory modalities ( Hainmueller and Bartos, 2020 ; Borzello et al., 2023 ). Despite its relatively simple trilaminar structure, the dentate gyrus has extensive molecular and cellular diversity ( Dyhrfjeld-Johnsen et al., 2007 ; Woods et al., 2018 ; Hainmueller and Bartos, 2020 ; Yao et al., 2021 ). SGCs are a dentate projection neuron subtype distinguished by sustained firing, wide dendritic arbors, and more frequent spontaneous inhibitory synaptic inputs than GCs ( Williams et al., 2007 ; Larimer and Strowbridge, 2010 ; Gupta et al., 2020 ; Afrasiabi et al., 2022 ). Here we demonstrate that SGCs receive more frequent spontaneous excitatory synaptic events and have a greater overall dendritic spine density than GCs. The higher SGC spine density was in proximal and middle dendritic segments which receive commissural/associational and MPP input. Correspondingly, the amplitude of MPP driven glutamatergic synaptic inputs to SGC was greater than in GCs. Similarly, the amplitude of EPSCs evoked by activation of the hilus was also greater in GCs indicating stronger associational input, as reported previously ( Williams et al., 2007 ). Unlike MPP inputs, the amplitude of EPSCs in response to LPP activation was not different between cell types. However, while LPP synapses to GCs showed short-term facilitation, LPP synapses to SGCs showed short-term depression suggesting that LEC activity may differentially recruit GCs and SGCs. We demonstrate that both SGCs and GCs receive commissural synaptic inputs from the contralateral dorsal and ventral hilus. Conversely, the strength and short-term dynamics of commissural synapses were not different between cell types. Thus, SGCs and GCs receive differential inputs from ipsilateral input streams with distinct information content while receiving similar contralateral inputs. Analysis of passive membrane models of SGCs and GCs reconstructed from biocytin fills revealed that cell-specific differences in dendritic morphology did not impact dendritic attenuation or synaptic integration between cell types. Moreover, SGCs were preferentially labeled as part of active neuronal ensembles both under basal conditions and during spatial memory task. Together, our data demonstrate differences in pathway specific input strength and short-term plasticity between SGCs and GCs which position SGCs to be preferentially activated by contextual information from the MEC and local associational inputs. Our demonstration that SGCs and GCs differ in the pathway and modality specific inputs from LEC and MEC is reminiscent of differential pathway-specific inputs to the superficial and deep CA1 pyramidal cells ( Masurkar et al., 2017 ). Indeed, as with CA1 deep pyramidal cells ( Bayer, 1980 ), SGCs are preferentially generated in an early embryonic phase and migrate further out radially to the inner molecular layer ( Save et al., 2019 ). The early born SGCs also receive stronger MEC inputs, as observed in deep CA1 pyramidal cells ( Masurkar et al., 2017 ). Additionally, SGCs receive stronger associational hilar inputs than GCs which parallels the difference in the strength of projections from CA2 to deep CA1 pyramidal cells ( Kohara et al., 2014 ). LEC inputs to GCs and SGCs differed, not in strength, but in short-term dynamics. Despite the layer-specific differences in ipsilateral inputs to SGCs and GCs, the strength and dynamics of commissural projections were not different between SGCs and GCs suggesting lateralization in processing of input streams. Our findings extend the known differences in morphology, intrinsic physiology, developmental timeline, and inhibition between SGCs and GCs ( Williams et al., 2007 ; Larimer and Strowbridge, 2010 ; Gupta et al., 2012 ; Save et al., 2019 ; Gupta et al., 2020 ; Afrasiabi et al., 2022 ) and identify diversity in glutamatergic inputs among dentate principal cells located along the radial axis. Whether SGCs and GCs differ in molecular and microcircuit features needs further investigation. Specifically, although SGCs were found to be enriched among dentate neurons expressing proenkephalin (PENK), an activity-dependent protein ( Erwin et al., 2020 ), PENK is expressed in multiple cell types including GCs ( Ziolkowska et al., 1998 ) and whether SGCs and GCs show distinct molecular profiles remains to be determined. Similarly, while recent reports indicate that glutamatergic synapses on dentate parvalbumin expressing interneurons are exclusively from SGCs ( Rovira-Esteban et al., 2020 ), whether there are functional differences in GC and SGC synaptic drive to dentate interneuron subtypes, as observed in CA1 pyramidal neurons ( Lee et al., 2014 ), needs to be evaluated. Apart from the subtypes of mature dentate projection neurons detailed above, the adult neurogenic niche with generation and maturation of GCs provides an added level of radial diversity in the dentate ( Altman and Das, 1965 ; Vivar et al., 2012 ). Immature GCs generated by adult neurogenesis are preferentially located in the inner third of the granule cell layer close to the hilus. Immature GCs are more excitable and receive stronger LEC synapses than mature GCs ( Vivar et al., 2012 ) indicating an additional layer of cellular complexity in the dentate circuit. Collectively, the dentate projection neuron types receiving different ipsilateral input streams with distinct information content are in a position to enhance the computational capacity of the dentate. Although the total dendritic length is not different between GCs and SGCs ( Afrasiabi et al., 2022 ), SGCs receive a greater frequency of spontaneous synaptic inputs and have a higher spine density than GCs. SGC spine density was higher than in GCs selectively in proximal and middle dendritic segments positioned to receive synaptic inputs from the MPP and associational pathways. Interestingly, the greatest difference between GCs and SGCs was in the higher density of thin dendritic spines, previously shown to be associated with learning and synaptic plasticity ( Popov et al., 2004 ; Bourne and Harris, 2007 ), in SGCs. These findings support SGCs engagement in memory function. Curiously, like mossy cells ( Buckmaster et al., 1993 ), SGCs have somatic spines and also exhibit spines on putative axons. Given the characteristic location of SGCs somata in the inner molecular layer, the somatic and axonal spines are ideally positioned to receive commissural and associational inputs from mossy cells. Mass activation of the perforant path elicited EPSCs with comparable amplitude in both GCs and SGCs, as previously reported in rats ( Williams et al., 2007 ). Since medial and lateral perforant path differ in information content and strength of glutamatergic inputs to mature and immature GCs ( Schmidt-Hieber et al., 2004 ; Vivar et al., 2012 ), we examined these specific input streams using pathway-specific ChR2 labeling. To control for the considerable variability in AAV driven reporter expression between animals, and even between sections, we adopted rigorous dual SGC-GC recordings at the same depth from the surface. Consistent with the lack of difference in distal dendritic spine density in GCs and SGCs, the amplitude of LPP driven EPSCs was not different between cell types. This contrasts with the stronger LPP driven EPSCs in adult-born immature GCs compared to mature GCs ( Vivar et al., 2012 ; Woods et al., 2018 ). Moreover, unlike GCs, SGCs exhibited short-term depression of a 10 Hz train of LPP evoked EPSCs which is distinct from the facilitation in LPP synapses observed in mature and immature GCs ( Vivar et al., 2012 ). Our study focused on mature GCs and SGCs explicitly and excluded immature GCs. Specifically, all GCs included in the study had somata in the middle and outer third of the granule cell layer, mature dendritic morphology and input resistance less than 400 MΩ consistent with mature GCs. Our data demonstrating target-specific differences in short-term dynamics of LPP synapses, is similar to reports in CA1 and cortical circuits ( Shigemoto et al., 1996 ; Thomson, 1997 ; Toth and McBain, 2000 ). Functionally, given the differences in short-term plasticity, it is likely that SGCs differ from GCs in summation and frequency-dependent filtering of LPP inputs ( Quintanilla et al., 2022 ). In contrast, MPP driven EPSCs were significantly stronger in SGCs than in GCs with no difference in multi-pulse depression. Correspondingly, SGCs had greater spine density than GCs in the proximal and middle dendritic segments which constitute the MEC input zone in SGCs. Once again, these data indicate that SGCs also differ from immature GCs which show significantly lower MPP driven EPSC amplitude than mature GCs and short-term facilitation rather than depression ( Vivar et al., 2012 ; Woods et al., 2018 ). Taken together, these data suggest that perforant path inputs differentially engage the SGCs, mature GCs and immature GCs for complex information processing. SGCs with strong MPP inputs and short-term depression are ideally suited to be activated by brief contextual information. Mature GCs receive facilitating LEC synapses and depressing MEC synapses which would better enable GCs to integrate sustained LEC activity with contextual information from MEC. Immature GCs receive large amplitude facilitating LEC synaptic inputs suggesting that they may strongly respond to object/content information. Because synaptic short-term dynamics can regulate information processing ( Tauffer and Kumar, 2021 ), SGCs are positioned to enhance the complexity of input processing in the dentate. Apart from entorhinal inputs, GCs receive commissural and associational glutamatergic projections from long-range mossy cell projections along the septotemporal axis ( Buzsàki and Eidelberg, 1981 ; Frotscher et al., 1991 ; Hsu et al., 2016 ; Botterill et al., 2021 ). Consistent with studies in rats ( Williams et al., 2007 ), hilar stimulation to activate local associational fibers elicited highly variable EPSCs in SGCs which were larger than in GCs. We find that both SGCs and mature GCs receive ventral and dorsal hilar commissural synapses with similar EPSC amplitude and depressing short-term dynamics. It is interesting to speculate that mossy cell synapses, especially associational projections, may target somatic spines on SGCs located in the IML. In this context mature and immature GCs also receive robust commissural inputs from contralateral mossy cells ( Chancey et al., 2014 ) with dorsal and ventral mossy cell axons showing differences in the distribution of their terminals in the dentate molecular layer ( Houser et al., 2020 ; Botterill et al., 2021 ). We also found extensive SGC axon collaterals in the IML (Supplementary Fig 4) which is consistent with SGC axon collaterals targeting parvalbumin expressing interneurons ( Rovira-Esteban et al., 2020 ). Although the IML collaterals could also contact GC dendrites, the lack of SGC to GC synapses in our recent paired recordings suggests that SGC to GC projections may not be extensive ( Dovek et al., 2024 ). How neurons integrate their inputs can be greatly influenced by dendritic structure including branching complexity, and passive and active properties ( Stuart and Spruston, 2015 ). Unlike hippocampal and cortical projection neurons, GCs show extensive branching of the proximal dendrite with limited subsequent branching. Morphologically constrained passive membrane models have identified considerably less distal dendritic EPSP attenuation in GCs than pyramidal cells ( Jaffe and Carnevale, 1999 ; Schmidt-Hieber et al., 2007 ), indicating that GC dendritic structure may be critical for input processing. Moreover, experimental and computational studies have demonstrated prominent EPSP attenuation at GC proximal dendrites ( Krueppel et al., 2011 ). Despite SGCs having lower dendritic complexity than GCs ( Gupta et al., 2020 ; Afrasiabi et al., 2022 ), our morphologically accurate passive membrane models revealed that dendritic EPSP attenuation and summation were not different between GCs and SGCs. However, the same peak synaptic conductance resulted in larger local dendritic EPSP in GCs than in SGCs, even though the models implemented identical channel densities, including leak conductance. It is possible that differences in membrane surface area in distal and middle dendritic segments of SGCs contribute to lower membrane resistance and contribute to the reduced local EPSP amplitude for the same synaptic conductance. Indeed, the larger size of the SGC somatic compartment than GC contributed to the lower input resistance in models SGC, which is consistent with experimental data ( Williams et al., 2007 ; Save et al., 2019 ; Afrasiabi et al., 2022 ; Dovek et al., 2024 ). Similarly, integration of coincident synaptic inputs within and between branches was sublinear and not different between cell types. The results in GCs are consistent with experimental data identifying that active dendritic properties including NMDA dependent calcium entry, sodium channels and t-type calcium channels, not included in our passive dendritic model, are needed to support the unique linear EPSP summation in GCs ( Krueppel et al., 2011 ). Further experimental characterization of dendritic active channels and synaptic integration in SGCs will be needed to determine whether SGCs, support linear summation or superliner summation of EPSPs on different dendritic branches as reported in GCs and cortical pyramidal cells, respectively ( Polsky et al., 2004 ; Krueppel et al., 2011 ). Regardless, our simulations demonstrate that while morphology impacts local EPSP amplitude, it does not differentially impact EPSP attenuation and summation in SGCs and GCs. Our evaluation of proportional SGC activation in task-naïve controls and mice trained in a spatial memory task revealed that about a quarter of the labeled neurons were SGCs, well above the 5% that SGCs represent among dentate GCs. The greater than expected labeling of SGCs as part of active neuronal ensembles in a spatial memory task is consistent with SGCs receiving stronger MEC inputs with spatial information and the disproportionate labeling of SGCs in a variety of contextual behavioral tasks ( Erwin et al., 2020 ; Dovek et al., 2024 ). The greater than expected SGC labeling among active neuronal ensembles in task-naive controls is unexpected. Since neurons are recruited to ensembles based on expression of the activity-dependent immediate early gene cFOS ( Yiu et al., 2014 ; Gouty-Colomer et al., 2016 ), these findings identify that SGCs have more basal activity than GCs, which aligns with their greater spontaneous glutamatergic drive and ability to support sustained firing. It is notable that the proportional labeling of SGCs was not different in task naïve and expert mice. This occurred despite the use of naïve littermate controls to avoid confounding effects of litter-to-litter variability in tdT expression and activity-dependent neuronal labeling in the TRAP2 mouse line. The proportional recruitment of SGCs in mice trained in spatial navigation and yoked task-naïve mice indicates that SGC activity scales in parallel with GC activity during a spatial memory task indicating task associated SGC activation during contextual memory formation. Our findings demonstrate layer and information modality-specific differences in glutamatergic inputs to GCs and SGCs extending the known differences between the two cell types. The evidence for dentate projection neuron types receiving different ipsilateral input streams with distinct information content indicates that SGCs and GCs are positioned to support parallel circuit processing of inputs and to enhance the computational capacity of the dentate. Materials and Methods STAR METHODS Lead Contact and Material Availability Further information and requests for resources and reagents listed in Table 1 should be directed to and will be fulfilled by the lead contact, Viji Santhakumar ( Vijayas{at}ucr.edu ). This study did not generate new unique reagents. View this table: View inline View popup Table 1: Key Resources Table Method Details Animals All experiments were conducted under IACUC protocols approved by the University of California at Riverside and conformed with ARRIVE guidelines. Male and female C57Bl6/J mice (p28-35), were used in slice electrophysiology studies. Inducible Fos-cre mice (TRAP2: Fostm2.1 (icre/ERT2)Luo/J ; Jackson Laboratories #030323) were back-crossed with C57BL6/N and were bred with reporter line tdT-Ai14 mice (B6;129S6-Gt(ROSA) 26Sortm14(CAG-tdTomato)Hze/J ; Jackson Laboratories # 007908) to create TRAP2-tdT mice for behavior experiments. Male and female TRAP2-tdT mice four to eight weeks old were used in behavior experiments. Mice were housed with littermates (up to 5 mice per cage) in a 12/12 h light/dark cycle. Food and water were provided ad libitum. Stereotaxic Injections Male and female C57Bl6/J mice were injected with AAV in lateral and medial entorhinal cortex (LEC & MEC) and hilus (dorsal and ventral) to selectively label specific glutamatergic inputs to GCs and SGCs. Juvenile mice (p28-35), were anesthetized with isoflurane and secured in small animal stereotaxic frame (Kopf). The scalp was injected (s.c) with bupivacaine (2mg/kg) at the site of incision. Animals also received a subcutaneous dose of carprofen (5mg/kg) and buprenorphine (Ethiqa XR at 3.25mg/kg) at start of surgery for analgesia. A 10µl Hamilton syringe was used with a Harvard Apparatus (11 Elite Nanomite) syringe pump for injections. Stereotaxic coordinates used for the entorhinal cortex regions: anterior/posterior (AP) in relation to true lambda; medial/lateral (ML) in relation to midline; and dorsal-ventral (DV) in relation to the dorsal brain surface. MEC: AP: -0.7, ML: 3.4, DV: -2.5. LEC: AP + 0.8, ML: 4.3, DV: -3.6 ( Donohue et al., 2021 ). To label MPP and LPP projections, mice were bilaterally injected with 180nl at 25nl/min of AAV-hSyn-hChR2(H134R)-EYFP (AAV5) or AAV-hSyn-EGFP (AAV5) for controls. Mice in which dorsal or ventral hilus were targeted, received unilateral injections of AAV-CaMKIIa-hChR2(H134R)-EYFP (AAV5) or AAV-CaMKIIa-EGFP (AAV5) for controls (300nl at 25nl/min) diluted in 1:1 sterile saline. The following hilar coordinates, AP is in relation to bregma were adopted: Dorsal hilus, AP: -2.1, ML: -1.25, DV: 1.9 from skull surface. Ventral hilus, AP: -3.4, ML: -2.7, DV: -3.4 from skull surface (adapted from Botterill et al., 2021 ). Injection site and projections with GFP and eYFP were visualized under epifluorescence illumination (λ=505 nm) using a Thorlabs 4-wave-length high-power LED source. Dual recordings from SGC-GC pairs were used to obtain optically evoked EPSCs. Dual recordings were essential to control for variable AAV labeling. pAAV-hSyn-hChR2(H134R)-EYFP and pAAV-CaMKIIa-hChR2(H134R)-EYFP were developed by Karl Deisseroth (Addgene viral prep # 26973-AAV5 & # 26969-AAV5; http://n2t.net/addgene:26973 & http://n2t.net/addgene:26969 ; RRID:Addgene_26973 & RRID:Addgene_26969, Lee et al., 2010 ; Chan et al., 2017 ) pAAV-hSyn-EGFP and pAAV-CaMKIIa-EGFP were developed by Bryan Roth (Addgene viral prep # 50465-AAV5 & Addgene viral prep # 50469-AAV5; http://n2t.net/addgene:50465 & http://n2t.net/addgene:50469 ; RRID:Addgene_50465 & RRID:Addgene_50469) Slice Physiology C57Bl6/J mice (p28-35), or C57Bl6/J mice 3-4 weeks following AAV injection were euthanized under isoflurane anesthesia for preparation of horizontal brain slices (300μm) using a Leica VT1200S Vibratome in ice cold sucrose artificial cerebrospinal fluid (sucrose-aCSF) containing (in mM): 85 NaCl, 75 sucrose, 24 NaHCO 3 , 25 glucose, 4 MgCl 2 , 2.5 KCl, 1.25 NaH 2 PO 4 , and 0.5 CaCl. Slices were bisected and incubated at 32°C for 30 min in a holding chamber containing an equal volume of sucrose-aCSF and recording aCSF and were subsequently held at room temperature for an additional 30 min before use. The recording aCSF contained (in mM): 126 NaCl, 2.5 KCl, 2 CaCl 2 , 2 MgCl 2 , 1.25 NaH 2 PO4, 26 NaHCO 3 , and 10 D-glucose. All solutions were saturated with 95% O 2 and 5% CO 2 and maintained at a pH of 7.4 for 2-6 hours ( Gupta et al., 2012 ; Yu et al., 2015 ; Afrasiabi et al., 2022 ). Slices were transferred to a submerged recording chamber and perfused with oxygenated aCSF at 33°C. Whole-cell voltage-clamp and current-clamp recordings from GCs in the granule cell layer and presumed SGCs in the inner molecular layer or edge of the granule cell layer were performed under IR-DIC visualization with Nikon Eclipse FN-1 (Nikon Corporation) using 40x water immersion objective. Recordings were obtained using filamented borosilicate microelectrodes (3-7MΩ) and recorded using Molecular Devices MultiClamp 700B amplifier. Data were low pass filtered at 2kHz, digitized using Axon DigiData 1400A (Molecular Devices) and acquired using pClamp11 at 10kHz sampling frequency. Recordings were performed using K-gluconate internal solution (K-gluc) containing 126 mM K-gluconate, 4 mM KCl, 10 mM HEPES, 4 mM Mg-ATP, 0.3 mM Na-GTP, and10 mM PO-creatinine or cesium methane sulfonate (CsMeSO4) internal solution (cesium) containing 140 mM cesium methane sulfonate, 10 mM HEPES, 5 mM NaCl, 0.2 mM EGTA, 2 mM Mg-ATP, and 0.2 mM Na-GTP. (pH 7.25; 270–290 mOsm). Biocytin (0.2%) was included in the internal solution for post hoc cell identification ( Gupta et al., 2012 ; Yu et al., 2015 ; Afrasiabi et al., 2022 ). Spontaneous and electrically evoked EPSCs were recorded using the cesium-based internal solution with cells held at -70mV in voltage clamp to isolate EPSCs. Evoked EPSCs were elicited by single stimulation of either the perforant path or the hilus using FHC stimulating electrodes placed at the junction of the dorsal blade and the crest just outside the fissure for PP ( Korgaonkar et al., 2020 ) or in the hilus avoiding the GCL and CA3 pyramidal cells for hilar stimulation. An Iso-Flex (AMPI Israel) stimulating box was used to deliver stimuli at 0.2 mA increments (3 sweeps). For experiments involving AAV-driven optically evoked EPSCs, SGC and GC pairs were recorded using a k-gluc based internal, were held at -70 mV in voltage clamp, and optically stimulated using wide-field blue (λ=470 nm 0.9mW) illumination using a Thorlabs 4-wave-length High Power LED source. Cells received a 10 ms, light pulse for 10 sweeps at 500 ms intervals or a 10 Hz train of light pulses with 10 ms pulse width. Recordings were discontinued if series resistance increased by > 20% or if access resistance was over 25MΩ. Post hoc biocytin immunostaining and morphological analysis were used to definitively identify all SGCs and GCs included in this study. Immunostaining, Cell Morphology and Reconstructions Following physiological recordings, slices were fixed in 0.1mM phosphate buffer containing 4% paraformaldehyde at 4°C overnight. Slices were washed with PBS and then incubated in Alexa Fluor® 594 conjugated streptavidin (1:1000 Thermo Fisher, S11227) in PBS with 0.3% Triton X-100 for 2 hours at room temperature. Following behavioral experiments, animals underwent transcardial perfusion with PBS followed by a 4% paraformaldehyde solution, under Euthasol anesthesia. Brains were extracted and stored in 4% paraformaldehyde at a temperature of 4°C for 3 hours before being transferred to PBS. The brains were sliced coronally into 50 µm sections using a Leica vt1000s vibratome; and 5 sequential sections were chosen, each 250 µm apart across the septotemporal axis. Free floating sections were blocked in 10% goat serum in PBS with 0.3% Triton X-100 for one hour. Sections were incubated in 4°C overnight in primary antibody for c-Fos (1:750, Rabbit mAb Cell Signaling Technology, cat #2250). The following day, sections were incubated in goat anti-rabbit Alexa Fluor® 488 secondary antibody (1:500 Abcam, cat #150077) for 1 hour and were then mounted on glass slides using Vectashield®.+DAPI mounting media ( Dovek et al., 2024 ). Sections were imaged using a Zeiss Axioscope-5 with stereo investigator (MBF Bioscience) for analysis. Cell counts and cell type classification and evaluation of double labeling were conducted by experimenter blinded to treatments. Cells with compact dendritic arbors and somata with greater length than width were classified as GCs and those with wide dendritic angle and 2 or more primary dendrites and greater somatic width than height were classified as SGCs ( Gupta et al., 2020 ; Afrasiabi et al., 2022 ; Dovek et al., 2024 ). Following posthoc immunostaining for biocytin recovery, a subset of biocytin filled cells were imaged using a Zeiss 800 Upright Microscope with Airyscan Fast using 20x lens at 0.23μm steps. Dendritic segments were imaged at higher resolution using a 40x oil lens at 0.04μm steps. Images were obtained at three regions along the cell: closest to the soma, the middle and end of the dendrites for proximal, middle and distal dendrites respectively. Spines were then reconstructed using Neurolucida 360. Reconstructions were user guided and began with initial tree construction in which all trees with complete imaging from all directions (checked through 3D rotation of the image) were reconstructed. Subsequently, using the user guided spine detection tool, each spine was selected by the user and reconstructed by the software. Automated spine classifications were used followed by experimenter verification. Counts and classifications were extracted using Neurolucida explorer. Due to variability in the length of dendritic segments imaged, spine densities were calculated by dividing the total number of spines from a segment by the total length of that segment. Simulations Granule cells and semilunar granule cells imaged using the Zeiss 800 Upright microscope were imported and reconstructed on Neurolucida 360. Following reconstructions, ASCII files were exported into “Trees” software ( Cuntz et al., 2011 ) to enable segmenting the dendrites into functionally distinct thirds representing the IML, MML and OML. Segmenting was performed following resampling to make distance between all nodes equal. Using “PVEC, PathLength” function, the dendrites were split into three segments and assigned titles of IML, MML, and OML accordingly. Segmented neuronal morphologies were imported into NEURON ( Hines and Carnevale, 2001 ) to insert conductance mechanisms. Active and passive conductance’s were distributed as described previously with a minor modification ( Table 2 ). GC and SGC somata were simulated as cylindrical compartments (in µm, GC: Length 16.8, Dia: 16.8; SGC: Length: 16.8, Dia 18.3) Dendrite channel densities were based on previously described parameters ( Santhakumar et al., 2005 ) and maintained identical in all dendritic compartments ( Table 2 ). View this table: View inline View popup Download powerpoint TABLE 2. Passive parameters and maximum conductance of the channels in model cells For experiments examining EPSP amplitude attenuation, a single, identical, AMPA synaptic conductance ( Golding et al., 2005 ) was located in the appropriate dendritic segment (IML, MML or OML) and activated with an artificial spike. The result in EPSP was measured both at the local dendrite and at the soma. Attenuation was evaluated by computing the EPSP peak amplitude at the soma as a percent of the local dendrite peak EPSP amplitude. A total of 9 trials with synapses located pseudo-randomly in different compartments within each of the three dendritic segments (IML,MML, and OML) were simulated in each reconstructed neuron. In experiments evaluating input summation, 5 AMPA synaptic conductances were randomly placed within the MML segments of the cell’s dendrites. Five independent simulations were conducted with synaptic conductance distributed either on the same branch or with the 5 synaptic conductances randomly distributed across branches in the middle dendritic segment. A total of 10 trials per reconstruction were implemented, and the peak EPSP amplitude at the soma compared between cell types. 4-Hydroxy Tamoxifen Preparation 4-Hydroxy Tamoxifen was dissolved in 100% ethanol at a concentration of 20mg/mL by sonicating solution at 37°C for 30 minutes or until it was fully dissolved. Solution was then aliquoted and stored at -20°C. On day of use, 4-OHT was redissolved by sonicating solution at 37°C for 10 minutes. A 1:4 mixture of castor oil and sunflower seed oil was added respectively to give a final concentration of 10mg/mL. The remaining ethanol in solution was evaporated by speed vacuuming in a centrifuge. Animals received 4-OHT (50mg/kg) intraperitoneally ( DeNardo et al., 2019 ; Dovek et al., 2024 ). Barnes Maze Male and female TRAP2-tdT mice were trained in a Barns maze spatial learning task. A cohort of litter mate controls went through all handling and activities other than being exposed to maze training. The Barnes maze table is 92cm in diameter and consists of 20 holes that are 5cm in diameter. A false floor installed with a removable escape box could be traded out for an additional false floor piece. The maze was set up in the middle of 4 curtain walls with two bright lights and a camera for recording above the maze. Attached to each curtain was a different set of visual cues (various shapes cut from felt). The escape hole was positioned in between two of the visual cues. Animals were kept outside of the curtain in a dark room until their turn to run the trial. Three days before testing, animals were handled for about a minute daily to habituate animals to experimenter. One day (Day 0) before beginning behavior, animals were individually housed due to the hyperactive nature of this particular mouse line and in order to prevent confounding results. Animals were brought to the behavior room and left alone to habituate in their cages for at least 1 hour before beginning training. Additional habituation was performed on day 1 of training, during which animals were placed in the starter cup on the table for 1 minute. They were then guided towards an escape box in a temporary location different from the experimental location. During the initial 1-6 days of acquisition training, each animal in the experimental group performed three, 180 second trials. For those animals in the active experimental group, the experimenter performed each trial for all the animals before starting the next trial with a minimum inter-trial-interval of 15 minutes. mice that did not enter the escape box at the end of the 180 seconds were guided by the experimenter to the escape box and placed back into their home cage. On day 6 of training, the animals were brought into the room 5 hours before testing. 10 minutes before Barnes maze testing, each mouse received 4-hydroxy tamoxifen (50mg/kg i.p). Animals were left in the room for an additional 5 hours after testing to limit neuronal activity labeling not related to the behavioral paradigm. During the probe trial on day 7, the escape box was replaced to look like all the other holes. The table was rotated 180 degrees to account for potential olfactory cues. Each animal was given 90 seconds to explore the table before being returned to the home cage. A week following induction (6 days following probe trial) on experimental day 12-13, animals repeated the training to reactivate the memory engram. Animals were sacrificed 90 minutes after performing the task on day 13 for immunostaining. In between each trial, the entire table and escape box were wiped down with 70% ethanol and allowed to fully dry. Behavioral data were analyzed using Any-maze software, by a blinded experimenter and using the automated BUNS analysis software ( Illouz et al., 2016 ). Quantification and Statistical Analysis Analysis of optical and stimulus evoked EPSCs as well as spontaneous EPSCs were conducted using EasyElectrophysiology software (Easy Electrophysiology Ltd) with a threshold search algorithm and events confirmed by the experimenter. Any “noise” that spuriously met trigger specifications was rejected. Spontaneous EPSCs were examined 180 seconds into the recording. Charge transfer (Area Under the curve-pA/ms), Amplitude, Rise Time (ms), and Biexponential Fit of Decay (ms) were calculated using the average trace of 100 events per cell. Instantaneous frequency was calculated from the first 100 accepted events. The reciprocal of inter-event interval (IEI) values generated in Easy Electrophysiology were used to compute instantaneous event frequencies used in further analysis. A bi-exponential fit was used when calculating decay. Rise time was calculated between 10 and 90% of the event amplitude. Sample sizes were not predetermined and conformed with those employed in the field. Significance was set to p<0.05, subject to appropriate Bonferroni Correction. Statistical analysis and figure generation were performed using GraphPad Prism 10. Unpaired Kolmogorov-Smirnov (KS) Test, unpaired Mann-Whitney, Nested-test, or one-way ANOVA followed by pairwise multiple comparisons using the Holm-Sidak method or Dunn’s method as appropriate. Data is presented as mean ± SEM. Additional graphics were created using Biorender. Author contributions Conceptualization, L.D and V.S.; Methodology, L.D and V.S.; Investigation, L.D., A.T. N, and E.G,; Writing – Original Draft, L.D and V. S; Writing – Review & Editing, L.D., A.T. N, E.G, V.S; Funding Acquisition, L.D and V.S; Supervision, L.D and V.S. Declaration of interests The authors declare no competing interests. Declaration of generative AI and AI-assisted technologies in the writing process No generative AI and AI-assisted technologies were used in writing this manuscript. Supplemental information Document S1. Figures S1–S5 Acknowledgements We thank Drs. Edward Zagha and Sachiko Haga-Yamanaka for thoughtful discussions. We thank Dr. David Carter at the UCR Microscopy Core and Mr. Erick Contreras for help with imaging. This work is supported by National Institutes of Health (NIH) NINDS R01NS069861, R01NS097750 to V.S., NIH/NINDS F31NS124290 to L.D, UCR RISE fellowship to E.G., AES BRIDGES to V.S., and E.G. Footnotes ↵ a Current Affiliation: Department of Neurology, Oregon Health & Science University References ↵ Afrasiabi M , Gupta A , Xu H , Swietek B , Santhakumar V ( 2022 ) Differential Activity-Dependent Increase in Synaptic Inhibition and Parvalbumin Interneuron Recruitment in Dentate Granule Cells and Semilunar Granule Cells . Journal of Neuroscience 42 : 1090 – 1103 . OpenUrl Abstract / FREE Full Text ↵ Altman J , Das GD ( 1965 ) Autoradiographic and histological evidence of postnatal hippocampal neurogenesis in rats . J Comp Neurol 124 : 319 – 335 . OpenUrl CrossRef PubMed Web of Science Aradi I , Holmes WR ( 1999 ) Role of multiple calcium and calcium-dependent conductances in regulation of hippocampal dentate granule cell excitability . J Comput Neurosci 6 : 215 – 235 . OpenUrl CrossRef PubMed Web of Science Aradi I , Soltesz I ( 2002 ) Modulation of network behaviour by changes in variance in interneuronal properties . J Physiol 538 : 227 – 251 . OpenUrl CrossRef PubMed Web of Science ↵ Bayer SA ( 1980 ) Development of the hippocampal region in the rat. I. Neurogenesis examined with 3H-thymidine autoradiography . J Comp Neurol 190 : 87 – 114 . OpenUrl CrossRef PubMed Web of Science ↵ Borzello M , Ramirez S , Treves A , Lee I , Scharfman H , Stark C , Knierim JJ , Rangel LM ( 2023 ) Assessments of dentate gyrus function: discoveries and debates . Nat Rev Neurosci 24 : 502 – 517 . OpenUrl CrossRef PubMed ↵ Botterill JJ , Gerencer KJ , Vinod KY , Alcantara-Gonzalez D , Scharfman HE ( 2021 ) Dorsal and ventral mossy cells differ in their axonal projections throughout the dentate gyrus of the mouse hippocampus . Hippocampus 31 : 522 – 539 . OpenUrl CrossRef PubMed ↵ Bourne J , Harris KM ( 2007 ) Do thin spines learn to be mushroom spines that remember? Current Opinion in Neurobiology 17 : 381 – 386 . OpenUrl CrossRef PubMed ↵ Buckmaster PS , Strowbridge BW , Schwartzkroin PA ( 1993 ) A comparison of rat hippocampal mossy cells and CA3c pyramidal cells . J Neurophysiol 70 : 1281 – 1299 . OpenUrl CrossRef PubMed Web of Science ↵ Buckmaster PS , Wenzel HJ , Kunkel DD , Schwartzkroin PA ( 1996 ) Axon arbors and synaptic connections of hippocampal mossy cells in the rat in vivo . J Comp Neurol 366 : 271 – 292 . OpenUrl CrossRef PubMed ↵ Buzsàki G , Eidelberg E ( 1981 ) Commissural projection to the dentate gyrus of the rat: evidence for feed-forward inhibition . Brain Research 230 : 346 – 350 . OpenUrl CrossRef PubMed Web of Science ↵ Chan KY , Jang MJ , Yoo BB , Greenbaum A , Ravi N , Wu WL , Sánchez-Guardado L , Lois C , Mazmanian SK , Deverman BE , Gradinaru V ( 2017 ) Engineered AAVs for efficient noninvasive gene delivery to the central and peripheral nervous systems . Nature Neuroscience 20 : 1172 – 1179 . OpenUrl CrossRef PubMed ↵ Chancey JH , Poulsen DJ , Wadiche JI , Overstreet-Wadiche L ( 2014 ) Hilar mossy cells provide the first glutamatergic synapses to adult-born dentate granule cells . J Neurosci 34 : 2349 – 2354 . OpenUrl Abstract / FREE Full Text ↵ Cuntz H , Forstner F , Borst A , Häusser M ( 2011 ) The TREES toolbox-probing the basis of axonal and dendritic branching . Neuroinformatics 9 : 91 – 96 . OpenUrl CrossRef PubMed Web of Science ↵ DeNardo LA , Liu CD , Allen WE , Adams EL , Friedmann D , Fu L , Guenthner CJ , Tessier-Lavigne M , Luo L ( 2019 ) Temporal evolution of cortical ensembles promoting remote memory retrieval . Nature Neuroscience 22 : 460 – 469 . OpenUrl CrossRef PubMed ↵ Donohue JD , Amidon RF , Murphy TR , Wong AJ , Liu ED , Saab L , King AJ , Pae H , Ajayi MT , Anderson GR ( 2021 ) Parahippocampal latrophilin-2 (ADGRL2) expression controls topographical presubiculum to entorhinal cortex circuit connectivity . Cell Reports 37 . ↵ Dovek L , Marrero K , Zagha E , Santhakumar V ( 2024 ) Cellular and circuit features distinguish dentate gyrus semilunar granule cells and granule cells activated during contextual memory formation . bioRxiv . ↵ Dyhrfjeld-Johnsen J , Santhakumar V , Morgan RJ , Huerta R , Tsimring L , Soltesz I ( 2007 ) Topological determinants of epileptogenesis in large-scale structural and functional models of the dentate gyrus derived from experimental data . JNeurophysiol 97 : 1566 – 1587 . OpenUrl CrossRef PubMed Web of Science ↵ Erwin SR , Sun W , Copeland M , Lindo S , Spruston N , Cembrowski MS ( 2020 ) A Sparse, Spatially Biased Subtype of Mature Granule Cell Dominates Recruitment in Hippocampal-Associated Behaviors . Cell Reports 31 : 107551 – 107551 . OpenUrl CrossRef PubMed ↵ Fernández-Ruiz A , Oliva A , Soula M , Rocha-Almeida F , Nagy GA , Martin-Vazquez G , Buzsáki G ( 2021 ) Gamma rhythm communication between entorhinal cortex and dentate gyrus neuronal assemblies . Science 372 . ↵ Fredes F , Silva MA , Koppensteiner P , Kobayashi K , Joesch M , Shigemoto R ( 2021 ) Ventro-dorsal Hippocampal Pathway Gates Novelty-Induced Contextual Memory Formation . Current Biology 31 : 25 – 38 .e25. OpenUrl CrossRef PubMed ↵ Frotscher M , Seress L , Schwerdtfeger WK , Buhl E ( 1991 ) The mossy cells of the fascia dentata: A comparative study of their fine structure and synaptic connections in rodents and primates . Journal of Comparative Neurology 312 : 145 – 163 . OpenUrl CrossRef PubMed Web of Science ↵ Golding NL , Mickus TJ , Katz Y , Kath WL , Spruston N ( 2005 ) Factors mediating powerful voltage attenuation along CA1 pyramidal neuron dendrites . Journal of Physiology 568 : 69 – 82 . OpenUrl CrossRef PubMed Web of Science ↵ Gouty-Colomer LA , Hosseini B , Marcelo IM , Schreiber J , Slump DE , Yamaguchi S , Houweling AR , Jaarsma D , Elgersma Y , Kushner SA ( 2016 ) Arc expression identifies the lateral amygdala fear memory trace . Molecular Psychiatry 21 : 364 – 375 . OpenUrl CrossRef PubMed ↵ Gupta A , Elgammal FS , Proddutur A , Shah S , Santhakumar V ( 2012 ) Decrease in tonic inhibition contributes to increase in dentate semilunar granule cell excitability after brain injury . Journal of Neuroscience 32 : 2523 – 2537 . OpenUrl Abstract / FREE Full Text ↵ Gupta A , Dovek L , Proddutur A , Elgammal FS , Santhakumar V ( 2022 ) Long-Term Effects of Moderate Concussive Brain Injury During Adolescence on Synaptic and Tonic GABA Currents in Dentate Granule Cells and Semilunar Granule Cells . Frontiers in Neuroscience 16 : 1 – 10 . OpenUrl ↵ Gupta A , Proddutur A , Chang YJ , Raturi V , Guevarra J , Shah Y , Elgammal FS , Santhakumar V ( 2020 ) Dendritic morphology and inhibitory regulation distinguish dentate semilunar granule cells from granule cells through distinct stages of postnatal development . Brain Structure and Function . ↵ Hainmueller T , Bartos M ( 2020 ) Dentate gyrus circuits for encoding, retrieval and discrimination of episodic memories . Nature Reviews Neuroscience 21 : 153 – 168 . OpenUrl CrossRef PubMed ↵ Hines ML , Carnevale NT ( 1997 ) The NEURON simulation environment . Neural Comput 9 : 1179 – 1209 . OpenUrl CrossRef PubMed Web of Science ↵ Hines ML , Carnevale NT ( 2001 ) NEURON: A tool for neuroscientists . Neuroscientist 7 : 123 – 135 . OpenUrl CrossRef PubMed Web of Science ↵ Houser CR , Peng Z , Wei X , Huang C , Mody I ( 2020 ) Mossy Cells in the Dorsal and Ventral Dentate Gyrus Differ in their Patterns of Axonal Projections . ↵ Hsu T-T , Lee C-T , Tai M-H , Lien C-C ( 2016 ) Differential Recruitment of Dentate Gyrus Interneuron Types by Commissural Versus Perforant Pathways . Cerebral Cortex 26 : 2715 – 2727 . OpenUrl CrossRef PubMed ↵ Illouz T , Madar R , Clague C , Griffioen KJ , Louzoun Y , Okun E ( 2016 ) Unbiased classification of spatial strategies in the Barnes maze . Bioinformatics 32 : 3314 – 3320 . OpenUrl CrossRef PubMed ↵ Jaffe DB , Carnevale NT ( 1999 ) Passive normalization of synaptic integration influenced by dendritic architecture . J Neurophysiol 82 : 3268 – 3285 . OpenUrl CrossRef PubMed Web of Science ↵ Knierim JJ , Lee I , Hargreaves EL ( 2006 ) Hippocampal place cells: Parallel input streams, subregional processing, and implications for episodic memory . Hippocampus 16 : 755 – 764 . OpenUrl CrossRef PubMed Web of Science ↵ Knierim JJ , Neunuebel JP , Deshmukh SS ( 2014 ) Functional correlates of the lateral and medial entorhinal cortex: objects, path integration and local–global reference frames . Philosophical Transactions of the Royal Society B: Biological Sciences 369 : 20130369 – 20130369 . OpenUrl CrossRef PubMed ↵ Kohara K , Pignatelli M , Rivest AJ , Jung HY , Kitamura T , Suh J , Frank D , Kajikawa K , Mise N , Obata Y , Wickersham IR , Tonegawa S ( 2014 ) Cell type-specific genetic and optogenetic tools reveal hippocampal CA2 circuits . Nat Neurosci 17 : 269 – 279 . OpenUrl CrossRef PubMed ↵ Korgaonkar AA , Li Y , Sekhar D , Subramanian D , Guevarra J , Swietek B , Pallottie A , Singh S , Kella K , Elkabes S , Santhakumar V ( 2020 ) Toll-like Receptor 4 Signaling in Neurons Enhances Calcium-Permeable α-Amino-3-Hydroxy-5-Methyl-4-Isoxazolepropionic Acid Receptor Currents and Drives Post-Traumatic Epileptogenesis . Annals of Neurology 87 : 497 – 515 . OpenUrl CrossRef PubMed ↵ Krueppel R , Remy S , Beck H ( 2011 ) Dendritic integration in hippocampal dentate granule cells . Neuron 71 : 512 – 528 . OpenUrl CrossRef PubMed Web of Science ↵ Larimer P , Strowbridge BW ( 2010 ) Representing information in cell assemblies: Persistent activity mediated by semilunar granule cells . Nature Neuroscience 13 : 213 – 222 . OpenUrl CrossRef PubMed Web of Science ↵ Lee JH , Durand R , Gradinaru V , Zhang F , Goshen I , Kim DS , Fenno LE , Ramakrishnan C , Deisseroth K ( 2010 ) Global and local fMRI signals driven by neurons defined optogenetically by type and wiring . Nature 465 : 788 – 792 . OpenUrl CrossRef PubMed Web of Science ↵ Lee SH , Marchionni I , Bezaire M , Varga C , Danielson N , Lovett-Barron M , Losonczy A , Soltesz I ( 2014 ) Parvalbumin-positive basket cells differentiate among hippocampal pyramidal cells . Neuron 82 : 1129 – 1144 . OpenUrl CrossRef PubMed Web of Science ↵ Masurkar AV , Srinivas KV , Brann DH , Warren R , Lowes DC , Siegelbaum SA ( 2017 ) Medial and Lateral Entorhinal Cortex Differentially Excite Deep versus Superficial CA1 Pyramidal Neurons . Cell Rep 18 : 148 – 160 . OpenUrl CrossRef PubMed Migliore M , Cook EP , Jaffe DB , Turner DA , Johnston D ( 1995 ) Computer simulations of morphologically reconstructed CA3 hippocampal neurons . J Neurophysiol 73 : 1157 – 1168 . OpenUrl CrossRef PubMed Web of Science ↵ Polsky A , Mel BW , Schiller J ( 2004 ) Computational subunits in thin dendrites of pyramidal cells . Nature Neuroscience 7 : 621 – 627 . OpenUrl CrossRef PubMed Web of Science ↵ Popov VI , Davies HA , Rogachevsky VV , Patrushev IV , Errington ML , Gabbott PL , Bliss TV , Stewart MG ( 2004 ) Remodelling of synaptic morphology but unchanged synaptic density during late phase long-term potentiation (LTP): a serial section electron micrograph study in the dentate gyrus in the anaesthetised rat . Neuroscience 128 : 251 – 262 . OpenUrl CrossRef PubMed Web of Science ↵ Quintanilla J , Jia Y , Lauterborn JC , Pruess BS , Le AA , Cox CD , Gall CM , Lynch G , Gunn BG ( 2022 ) Novel types of frequency filtering in the lateral perforant path projections to dentate gyrus . J Physiol 600 : 3865 – 3896 . OpenUrl CrossRef PubMed ↵ Rovira-Esteban L , Hajos N , Nagy GA , Crespo C , Nacher J , Varea E , Blasco-Ibanez JM ( 2020 ) Semilunar Granule Cells Are the Primary Source of the Perisomatic Excitatory Innervation onto Parvalbumin-Expressing Interneurons in the Dentate Gyrus . eNeuro 7 . ↵ Santhakumar V , Aradi I , Soltesz I ( 2005 ) Role of mossy fiber sprouting and mossy cell loss in hyperexcitability: A network model of the dentate gyrus incorporating cell types and axonal topography . Journal of Neurophysiology 93 : 437 – 453 . OpenUrl CrossRef PubMed Web of Science ↵ Save L , Baude A , Cossart R ( 2019 ) Temporal Embryonic Origin Critically Determines Cellular Physiology in the Dentate Gyrus . Cerebral Cortex 29 : 2639 – 2652 . OpenUrl CrossRef PubMed ↵ Schmidt-Hieber C , Jonas P , Bischofberger J ( 2004 ) Enhanced synaptic plasticity in newly generated granule cells of the adult hippocampus . Nature 429 : 184 – 187 . OpenUrl CrossRef PubMed Web of Science ↵ Schmidt-Hieber C , Jonas P , Bischofberger J ( 2007 ) Subthreshold dendritic signal processing and coincidence detection in dentate gyrus granule cells . J Neurosci 27 : 8430 – 8441 . OpenUrl Abstract / FREE Full Text ↵ Shigemoto R , Kulik A , Roberts JD , Ohishi H , Nusser Z , Kaneko T , Somogyi P ( 1996 ) Target-cell-specific concentration of a metabotropic glutamate receptor in the presynaptic active zone . Nature 381 : 523 – 525 . OpenUrl CrossRef PubMed Web of Science ↵ Slomianka L , Amrein I , Knuesel I , Sorensen JC , Wolfer DP ( 2011 ) Hippocampal pyramidal cells: the reemergence of cortical lamination . Brain Struct Funct 216 : 301 – 317 . OpenUrl CrossRef PubMed ↵ Soltesz I , Losonczy A ( 2018 ) CA1 pyramidal cell diversity enabling parallel information processing in the hippocampus . Nat Neurosci 21 : 484 – 493 . OpenUrl CrossRef PubMed ↵ Stuart GJ , Spruston N ( 2015 ) Dendritic integration: 60 years of progress . Nature Neuroscience 18 : 1713 – 1721 . OpenUrl CrossRef PubMed ↵ Tauffer L , Kumar A ( 2021 ) Short-Term Synaptic Plasticity Makes Neurons Sensitive to the Distribution of Presynaptic Population Firing Rates . eNeuro 8 . ↵ Thomson AM ( 1997 ) Activity-dependent properties of synaptic transmission at two classes of connections made by rat neocortical pyramidal axons in vitro . J Physiol 502 ( Pt 1 ): 131 – 147 . OpenUrl CrossRef PubMed Web of Science ↵ Toth K , McBain CJ ( 2000 ) Target-specific expression of pre- and postsynaptic mechanisms . J Physiol 525 Pt 1 : 41 – 51 . OpenUrl CrossRef PubMed Web of Science ↵ Vivar C , Potter MC , Choi J , Lee JY , Stringer TP , Callaway EM , Gage FH , Suh H , van Praag H ( 2012 ) Monosynaptic inputs to new neurons in the dentate gyrus . Nat Commun 3 : 1107 . OpenUrl CrossRef PubMed ↵ Williams PA , Larimer P , Gao Y , Strowbridge BW ( 2007 ) Semilunar granule cells: Glutamatergic neurons in the rat dentate gyrus with axon collaterals in the inner molecular layer . Journal of Neuroscience 27 : 13756 – 13761 . OpenUrl Abstract / FREE Full Text ↵ Woods NI , Vaaga CE , Chatzi C , Adelson JD , Collie MF , Perederiy JV , Tovar KR , Westbrook GL ( 2018 ) Preferential targeting of lateral entorhinal inputs onto newly integrated granule cells . Journal of Neuroscience 38 : 5843 – 5853 . OpenUrl Abstract / FREE Full Text ↵ Yao Z et al. ( 2021 ) A taxonomy of transcriptomic cell types across the isocortex and hippocampal formation . Cell 184 : 3222 – 3241 e3226. OpenUrl CrossRef PubMed ↵ Yiu AP , Mercaldo V , Yan C , Richards B , Rashid AJ , Hsiang HLL , Pressey J , Mahadevan V , Tran MM , Kushner SA , Woodin MA , Frankland PW , Josselyn SA ( 2014 ) Neurons Are Recruited to a Memory Trace Based on Relative Neuronal Excitability Immediately before Training . Neuron 83 : 722 – 735 . OpenUrl CrossRef PubMed Web of Science ↵ Yu J , Swietek B , Proddutur A , Santhakumar V ( 2015 ) Dentate total molecular layer interneurons mediate cannabinoid-sensitive inhibition . Hippocampus 25 : 884 – 889 . OpenUrl CrossRef PubMed ↵ Ziolkowska B , Przewlocka B , Mika J , Labuz D , Przewlocki R ( 1998 ) Evidence for Fos involvement in the regulation of proenkephalin and prodynorphin gene expression in the rat hippocampus . Brain Res Mol Brain Res 54 : 243 – 251 . OpenUrl CrossRef PubMed View the discussion thread. Back to top Previous Next Posted March 15, 2025. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Differential Glutamatergic Inputs to Semilunar Granule Cells and Granule Cells Underscore Dentate Gyrus Projection Neuron Diversity Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Differential Glutamatergic Inputs to Semilunar Granule Cells and Granule Cells Underscore Dentate Gyrus Projection Neuron Diversity Laura Dovek , Anh-Tho Nguyen , Emmanuel Green , Vijayalakshmi Santhakumar bioRxiv 2025.03.14.643192; doi: https://doi.org/10.1101/2025.03.14.643192 Share This Article: Copy Citation Tools Differential Glutamatergic Inputs to Semilunar Granule Cells and Granule Cells Underscore Dentate Gyrus Projection Neuron Diversity Laura Dovek , Anh-Tho Nguyen , Emmanuel Green , Vijayalakshmi Santhakumar bioRxiv 2025.03.14.643192; doi: https://doi.org/10.1101/2025.03.14.643192 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Neuroscience Subject Areas All Articles Animal Behavior and Cognition (7635) Biochemistry (17691) Bioengineering (13892) Bioinformatics (41937) Biophysics (21452) Cancer Biology (18588) Cell Biology (25504) Clinical Trials (138) Developmental Biology (13378) Ecology (19899) Epidemiology (2067) Evolutionary Biology (24320) Genetics (15609) Genomics (22506) Immunology (17736) Microbiology (40394) Molecular Biology (17181) Neuroscience (88605) Paleontology (666) Pathology (2832) Pharmacology and Toxicology (4824) Physiology (7641) Plant Biology (15156) Scientific Communication and Education (2045) Synthetic Biology (4294) Systems Biology (9825) Zoology (2271)