Development of a floxed Gabbr2 gene allows for widespread conditional disruption of GABBR2 and recapitulates the phenotype of germline Gabbr2 knockout mice

1 1 Development of a floxed Gabbr2 gene allows for widespread conditional disruption of 2 GABBR2 and recapitulates the phenotype of germline Gabbr2 knockout mice. 3 4 Julie R. Hens1*, Stacey Brown1, Pawel Licznerski1, Jacqueline Suarez1, Elizabeth Jonas1, and 5 John J. Wysolmerski1. 6 7 1Department of Internal Medicine, Endocrinology and Metabolism Section, Yale University, 8 New Haven, Connecticut, United States of America. 9 10 * Corresponding author 11 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 2 12 ABSTRACT 13 GABBR1 and GABBR2 are widely expressed in the brain and genetic inhibition of their 14 function leads to widespread neurologic dysfunction and premature death in mice. Given 15 that GABBR1 and GABBR2 heterodimerize to form a functional receptor, global knockout of 16 GABBR1 or GABBR2 results in a similar phenotype, characterized by spontaneous 17 epileptiform activity, hyperlocomotor activity, hyperalgesia, impaired memory and 18 premature death. It is now known that both GABBR1 and GABBR2 are expressed in a 19 variety of tissues outside the nervous system and that GABA-B receptors can 20 heterodimerize with other class C GPCRs, including the extracellular calcium-sensing 21 receptor (CaSR). Studies in vitro have demonstrated that interactions with GABBR1 and 22 GABBR2 can alter CaSR signaling in human embryonic kidney cells and breast cancer cells. 23 The neurologic consequences of global loss of function of GABBR1 or GABBR2 has made it 24 difficult to study the effects of loss of GABBR function in other organs. While a conditional 25 knockout for GABBR1 is available, the GABBR2 gene had not been “floxed”. We have used 26 CRISPR to insert loxP sites into the GABBR2 locus in mice. These mice are normal at 27 baseline but when bred with mice expressing Cre-recombinase under the control of the 28 ubiquitously expressed Actin gene promoter, they recapitulate the phenotype of global 29 GABBR2 knockout mice. Phenotypic changes through the brain, including the cortex, 30 hippocampus and cerebellum. Evidence of abnormal neuronal function, increase cell death, 31 and changes in neuronal architecture are seen throughout the brain of CRISPR knockout 32 mice. These mice should be useful tools to study cell type-specific loss of GABBR2 function 33 in the brain and other organs. 34 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 3 35 INTRODUCTION 36 The GABA-B receptors (GABBR1 and GABBR2) are class C, G-protein-coupled 37 receptors (GPCRs) that heterodimerize to form a receptor complex responding to gamma- 38 aminobutyric acid (gaba), the major inhibitory neurotransmitter in the brain. The GABBR1 39 subunit contains the gaba-binding site, whereas the GABBR2 subunit is responsible for 40 interacting with G-proteins. Furthermore, GABBR1 contains an endoplasmic reticulum 41 (ER) retention site, which prevents its trafficking to the plasma membrane. However, 42 heterodimerization with GABBR2 allows interactions between the coiled-coil sequences of 43 each subunit, masking the ER retention site in GABBR1, and allowing translocation of the 44 heterodimeric complex to the plasma membrane. Most commonly, the heterodimeric 45 receptor couples to Gi or Go, leading to inhibition of adenylate cyclase activity, inositol 46 triphosphate synthesis, voltage-gated calcium channels, and potassium channels (1, 2). As 47 a result, GABABRs hyperpolarize neurons and inhibit the release of several 48 neurotransmitters, resulting in the suppression of neuronal activity in many brain areas. 49 GABBR1 and GABBR2 are widely expressed in the brain and genetic inhibition of 50 their function leads to widespread neurologic dysfunction and premature death in mice. 51 Given that GABBR1 and GABBR2 heterodimerize to form a functional receptor, global 52 knockout of GABBR1 or GABBR2 results in a similar phenotype, characterized by 53 spontaneous epileptiform activity, hyperlocomotor activity, hyperalgesia, impaired 54 memory and premature death (3). As these results demonstrate, GABA-B receptors clearly 55 have important functions in the brain. However, it is now known that both GABBR1 and 56 GABBR2 are expressed in a variety of tissues outside the nervous system (3-6). 57 Furthermore, it has been shown that the GABA-B receptors can heterodimerize with other .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 4 58 class C GPCRs, including the extracellular calcium-sensing receptor (CaSR)(7, 8). Studies in 59 vitro have demonstrated that interactions with GABBR1 and GABBR2 can alter CaSR 60 signaling in human embryonic kidney (HEK) cells and breast cancer cells (5, 9). 61 Furthermore, the CASR and GABBR1 interact in chondrocytes in the growth plate and in 62 parathyroid cells in vivo (4). Recent studies have demonstrated that heterodimerization of 63 GABBR1 and the CaSR in the parathyroid glands modulates calcium-mediated PTH 64 secretion and systemic calcium metabolism (10), demonstrating that GABBR’s can regulate 65 signaling from other receptors. 66 The neurologic consequences of global loss of function of GABBR1 or GABBR2 has 67 made it difficult to study the effects of loss of GABBR function in other organs. While a 68 conditional knockout for GABBR1 is available, the GABBR2 gene had not been “floxed”. 69 Therefore, to study the interactions between the CaSR and GABBR2 in organs other than 70 the brain, we have used gene editing techniques to insert loxP sites into the GABBR2 locus 71 in mice. These mice are normal at baseline but when crossed with mice expressing Cre- 72 recombinase under the control of the ubiquitously expressed Actin gene promoter, they 73 recapitulate the phenotype of global GABBR2 knockout mice. These mice should be useful 74 tools to study cell type-specific loss of GABBR2 function in the brain and other organs. 75 76 77 METHODS 78 Generation and breeding of Gabbr2 cKO Mice 79 The GABBR2 cKO mouse model was generated via CRISPR-Cas9 genome editing (11, 12) 80 (13). Potential Cas9 target guide (protospacer) sequences in introns 9 and 10 were .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 5 81 screened using the online tool CRISPOR http://crispor.tefor.net (14) and candidates were 82 selected. Templates for sgRNA synthesis were generated by PCR, sgRNAs were transcribed 83 in vitro and purified (Megashortscript, MegaClear; ThermoFisher). sgRNA/Cas9 RNPs were 84 complexed and tested for activity by zygote electroporation, incubation of embryos to 85 blastocyst stage, and genotype scoring of indel creation at the target sites. The sgRNAs that 86 demonstrated the highest activity were selected for creating the floxed allele. Guide RNA 87 (gRNA) sequences are as follows: intron 9, 5’ guide: ACTAGATCCTCTCACCCAGT and intron 88 10, 3’ guide CTGCCATGCTGTGACCCCAT. Accordingly, a 615 base long single-stranded DNA 89 (lssDNA) recombination template incorporating the 5’ and 3’ loxP sites was synthesized 90 (IDT). The C57Bl6 3 SJL F2 or FVB/NJ zygote embryos were transferred to the oviducts of 91 pseudopregnant CD-1 foster females using standard techniques(13, 15). Genotype 92 screening of tissue biopsies from founder pups was performed by PCR amplification and 93 Sanger sequencing to verify the floxed allele. Germline transmission of the correctly 94 targeted allele (i.e., both loxP sites in cis) was confirmed by breeding and sequence analysis. 95 Seven potential founders with a floxed Gabbr2 gene were identified, and three (#33, #14, 96 #19) true-breeding FVB lines were generated. We also generated two true-breeding lines 97 on a C57bl/6 mouse background. The studies described herein were performed on animals 98 derived from lines 33 and 19. The two lines were maintained separately, but because of 99 their similar biochemical phenotypes, data from the two lines have been pooled except 100 where indicated. 101 We crossed GabbR2 lox/lox mice with B6.FVB-Tmem163Tg(ACTB-cre)2Mrt/EmsJ (Actin-cre) to 102 verify the effectiveness of the CRISPR-generated lox sites on GabbR2. The resulting .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 6 103 offspring were then crossed again to GABBR2 lox/lox mice (control) to generate Actin-Cre/ 104 GABBR2 lox/lox mice (cKO). 105 All studies described in this manuscript were performed on animals between the 106 ages of 3 and 15 weeks unless specifically indicated. All procedures were per Yale 107 University Animal Care and Use Committee and U.S. National Institutes of Health standards. 108 109 RNA and protein analysis 110 Brains from 3-week-old mice were removed and total RNA was isolated. One g of 111 RNA was converted to cDNA using Applied Biosystems high-capacity cDNA reverse 112 transcription kit (Thermo Fisher Scientific, Waltham, MA). Taqman probes were used to 113 measure GABBR1(Mm00444578_m1), GABBR2(Mm01352554_m1), CASR 114 (Mm00443375_m1), and GAPDH (#4352339E) (Thermo Fisher Scientific, Waltham, MA). 115 Real-time PCR was performed using TaqMan TM Fast Universal PCR Master Mix reagents 116 (Thermo Fisher Scientific, Waltham, MA) and Applied Biosystems StepOne Plus Real-Time 117 PCR System. Ct values were analyzed using the ΔΔ – Ct method (16). 118 For protein isolation, half the brain cut in the coronal mid-line was added to 1 ml of 119 RIPA buffer (50 mM Tris HCl pH 8, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 120 0.1% SDS) with complete mini protease inhibitors (Roche Diagnostics, Mannheim, 121 Germany). Using a TissueLyser II (Qiagen, Germantown, MD) with a 5 mm bead, tissue was 122 lysed for 2 minutes at 30 rotations per second. Lysates were incubated on ice for an hour, 123 before being centrifuged at 12,000 g, for 20 minutes. Thirty micrograms of protein were 124 loaded in a well. Samples were not heated, and after the transfer, blots were blocked for 1 125 hour in 5% milk with 0.1 % Tween-20. Primary antibodies were added at 1/1000 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 7 126 overnight at 40C while rocking. Blots were washed with PBS, and then goat-anti rabbit or 127 goat-anti mouse secondary antibody was added for an hour, samples were washed in PBS. 128 Blots were imaged using Odyssey Li-Cor system. Results were normalized to actin. 129 We used antibodies to Gabbr1 (#ab55051, Abcam, Waltham, MA), Gabbr2 130 (#ab181736, Abcam, Waltham, MA), Casr (#ACR-004, Alomone, Limerick, PA), actin 131 (#MA5-11869, Invitrogen, Rockford Illinois) , IRDye® 800CW Goat anti-Mouse IgG (H + L) 132 (Li-Cor, Lincoln, Nebraska) IRDye®, 680RD Goat anti-Rabbit IgG Secondary Antibody (Li- 133 Cor, Lincoln, Nebraska) 134 135 Histology 136 Brains were paraffin-embedded and 5-micron sections were acquired. Sections 137 were stained with hematoxylin and eosin, Luxol fast (17), or immunohistochemistry was 138 performed with S100 antibody to examine changes in myelination in the central nervous 139 system. Embedding and staining of mice brain tissue was done through Yale Pathology 140 Tissue services. 141 142 Motor agility 143 To examine motor changes in cKO compared to control mice, we assessed rotarod 144 performance. Mice were trained to stay on the rotarod (AccuScan Instruments) (12 rpm) 145 for 300 sec over two separate sessions the day before the experiment. During the test day, 146 the length of time each mouse remained on the cylinder (“endurance time”; a maximal 147 score of 300 sec) was measured immediately before (time 0) and 1, 2, and 4 hours after the 148 application of L-baclofen (12.5 mg/kg) or vehicle (saline). The dose of baclofen that .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 8 149 showed maximal effects on rotarod performance was determined in previous studies (3, 150 18). 151 152 Behavioral experiments 153 We examined hyperactivity and unsupported rearing to analyze activity in the mice. 154 Male cKO mice between 6 and 8 weeks of age were used for all experiments. Before 155 behavioral testing, the investigator individually handled mice (3 times over 72 hours 156 before the test day) to decrease anxiety. Next, mice were placed in a new, empty home cage 157 where unsupported rearing and locomotor activity were monitored for 10-minute sessions, 158 video recorded, and the last 5 minutes were scored manually. Unsupported rearing was 159 defined as rearing without any contact with the walls of the test cage. The investigator was 160 blinded as to the genetic variant during scoring. 161 162 Statistical analysis 163 Data are presented as mean± standard error (SE). Comparisons between two groups 164 were conducted using Student's unpaired two-tailed t-tests. Where appropriate, two-way 165 ANOVA with Sidak multiple comparison tests were used. All analyses were performed 166 using Prism 10 (GraphPad Software, La Jolla, CA). 167 168 169 RESULTS 170 Insertion of LoxP sites and reduction in GABBR2 expression. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 9 171 Using CRISPR we inserted loxP sites into 5’ and 3’ sites flanking exon 10 of the 172 Gabbr2 gene, which encodes the first transmembrane domain of the receptor (Figure 1a). 173 We targeted this exon for several reasons. First, it was predicted to result in the loss of the 174 first transmembrane domain. Second excision of this portion of DNA was predicted to 175 result in a frameshift and mistranslation of all downstream exons when the primary 176 transcript was spliced. Both of these characteristics are likely to result in a nonfunctional 177 protein that would be degraded. Finally, targeting this relatively small exon allowed both 178 flanking loxP sites to be targeted with one oligomer, allowing for more efficient editing. 179 Primers were designed to detect wild-type and loxP sites at the 5’ and 3’ end of exon 10 to 180 detect the appropriately floxed alleles (Figure 1b). Using these primers, we identified 5 181 potential founder lines in a FVB background that contained both loxP sites, three of which 182 passed on the correct allele in a Mendelian fashion. We also identified 2 founder lines in a 183 C57Bl/6 background, both of which passed on the correct genotype to offspring in 184 Mendelian fashion (Figure 1c). We used lines 19 and 33 in an FVB background, (referred to 185 as Gabbr2lox/lox mice) in the following experiments. 186 Gabbr2lox/lox mice were bred to B6.FVB-Tmem163Tg(ACTB-cre)2Mrt/EmsJ (Actin-cre) 187 mice to generate Gabbr2 cKO mice with widespread loss of GABBR2 expression. In order to 188 verify the loss of GABBR2, we examined Gabbr2 mRNA levels in whole brains from 21-day- 189 old mice. Gabbr2 mRNA expression was reduced by 80% in the Gabbr2 cKO mice as 190 compared to Gabbr2lox/lox (control) mice, lacking Cre expression. Loss of Gabbr2 mRNA 191 expression did not affect either Gabbr1mRNA or Casr mRNA levels, two potential 192 heterodimerization partners for GABBR2 (Figure 2A). We assessed GABBR2 expression by 193 immunoblots of whole brain extracts. As shown in Fig. 2B, no GABBR2 protein was .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 10 194 detected in extracts of whole brains harvested from cKO mice, although it was easily 195 detected in brain extracts from control mice. As with the mRNA levels, loss of GABBR2 196 protein did not affect GABBR1 or CASR protein levels (Fig. 2B). These results demonstrate 197 the effective elimination of GABBR2 expression when Gabbr2lox/lox mice are bred with Cre 198 recombinase-expressing mice. 199 200 Histological changes in the brain due to the loss of GABBR2 201 GABBR2 is expressed throughout the brain, including the cerebral cortex, 202 cerebellum, Purkinje neurons, hippocampus, CA3 neurons, thalamic nuclei, medial 203 habenula, and astrocytes (19-23). S100 proteins are expressed diffusely in glial cells, 204 astrocytes and neurons throughout the brain (24, 25). In the GABBR2 cKO cortex, there 205 was a generalized decrease in diffuse S100 staining and fewer distinct S100-positive cells 206 when compared to control mice (Figure 3A versus 3B, red arrows). In addition, there were 207 fewer S100-positive dendritic extensions in the GABBR2 cKO cortex (Figure 3A versus 3B, 208 yellow arrows). There was also an increase in vacuolated neuronal bodies and cell debris 209 evident in Luxol blue stained sections (Figure 3C versus 3D, green arrows), suggesting 210 potential neuronal damage. 211 Changes were also evident in the dentate gyrus and CA3 region of the hippocampus 212 of GABBR2 cKO mice. There was a clear reduction in staining of the CA3 region (Figure 4A 213 versus 4B, blue arrows). We observed a clear reduction in the number and layers of dense 214 immature granular cells (Figure 4A versus 4B, green arrows and dotted border) as well as 215 an increase in vacuolated cytoplasm in granular cells (Figure 4A versus 4B, and Figure 4C .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 11 216 versus 4D, red arrows). There were many shrunken pyramidal cells in the polymorphic 217 cell layer (Figure 4A and 4B, yellow arrows). 218 Finally, the cerebellum of GABBR2 cKO mice demonstrated alterations in the 219 organization of the Purkinje cell layer, with fewer Purkinje neurons, and more swollen or 220 vacuolated cells (Figure 5A versus 5B, yellow arrows). Additionally, there were fewer 221 dendritic projections penetrating into the molecular layer and reduced complexity of the 222 dendritic branching pattern. (Figure 5A versus Figure 5B, red arrows). 223 224 Loss of GABBR2 Alters Behavior and Motor Skills 225 Previous reports on the global GABBR2 KO mice described hyperalgesia, 226 hyperlocomotion, elevated anxiety-related behaviors, and spontaneous seizure activity (3, 227 26). Therefore, we examined these activities in GABBR2 cKO mice to determine whether 228 they mimicked the phenotype of global GABBR2 KO Mice. GABBR2 cKO mice 229 demonstrated a greater than 3-fold increase in locomotor activity compared to control 230 mice (Figure 6). There was a significant reduction of unsupported rearing behavior in 231 GABBR2 cKO mice as compared to controls (Figure 6). This decrease in exploratory 232 behavior is likely indicative of increased levels of stress but can also be seen in the setting 233 of neurodegenerative disorders (27-29). 234 Baclofen is an agonist for gamma-aminobutyric acid (GABA) B receptors, and acts as 235 a muscle relaxant (30, 31). Global GABBR2 knockout mice were previously shown to be 236 refractory to baclofen as measured by changes in rotarod performance (3). Therefore, we 237 assessed rotarod performance and responses to baclofen in GABBR2 cKO and control mice. 238 During the rotarod training period preceding baclofen administration, it was clear that cKO .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 12 239 mice of both sexes had a baseline decrease in their ability to remain on the rotarod (Figure 240 7A). Therefore, we expressed the response to baclofen as the change from baseline. 241 Control mice of both sexes had a clear decrease in rotarod performance after baclofen 242 treatment. However, despite the reduced performance at baseline, cKO mice showed no 243 additional decline in performance after administration of baclofen (Figure 7B). 244 245 Loss of GABBR2 results in seizures and premature death. 246 We did not detect obvious spontaneous seizure activity while GABBR2 cKO mice 247 were being monitored for locomotor activity. However, these mice had frequent seizures 248 when subjected to stressful stimuli, such as, being handled or placed on the rotarod 249 (Supplemental Video). In addition, we noted an increase in premature mortality in 250 GABBR2 cKO mice. As shown in Figure 8, 100% of GABBR2 cKO mice died by 115 days of 251 age while no control mice died during the same period. In addition to the pathological 252 brain findings described above, necropsy of GABBR2 cKO mice showed little food in the 253 stomach and small intestines, but no gross pathological changes. However, there was 254 marked thymic necrosis. The spleen was enlarged and had areas of lymphocytic necrosis. 255 The pancreas showed an absence of eosinophilic zymogen granules within the exocrine 256 pancreatic acinar cells. Mice that had died some time before necropsy had brain findings 257 similar to those described above. There was mild dilation of the lateral ventricles, multiple 258 foci where there was decreased staining of the neuropil, especially in areas where cell 259 nuclei were shrunken and there was cytoplasmic vacuolar degeneration. Diffuse 260 demyelination was seen, blood vessels appeared congested, and some vessels contained 261 mature fibrin. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 13 262 263 Discussion: 264 We generated a floxed GABBR2 mouse, that when crossed with Cre-expressing mice 265 can be used to generate conditional KO mice targeting GABBR2 expression in different cell 266 types. In this report, we crossed floxed GABBR2 mice with Actin-Cre mice to produce a 267 widespread knockout of the Gabbr2 gene. Gabbr2 mRNA expression was reduced by over 268 80% in the whole brain and GABBR2 protein was not detected by immunoblot, 269 documenting efficient disruption of the Gabbr2 gene. The phenotype of Actin-Cre GABBR2 270 cKO mice was similar to the global GABBR2 KO mouse (3). These mice demonstrated 271 increased locomotor activity but a decrease in unsupported rearing behavior. These mice 272 also have impaired motor coordination and balance as measured by decreased ability to 273 remain on a rotarod and seizures in response to being handled. These neuro-behavioral 274 changes were accompanied by widespread changes in brain histology and also a reduced 275 lifespan, both speaking to the importance of GABBR2 signaling for overall brain health and, 276 perhaps, whole-body physiology as well. 277 We found that the loss of GABBR2 led to histological changes in different areas of 278 the brain. In the cortex, GABBR2 is expressed in many different neurons including 279 GABAergic cortical interneurons (32) and inhibitory interneurons (33). Loss of GABBR2 280 would be expected to impair slow inhibitory Gaba signaling to the interneurons connecting 281 different regions of the cortex, perhaps resulting in a progressive decline in interneuron 282 function. Loss of inhibitory interneuron signaling may also result in changes in cell 283 viability as reflected here as a reduction of cortical thickness, a decrease in S100 staining, 284 reductions in dendritic extensions and the presence of vacuolated neurons and cellular .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 14 285 debris. Functionally, such a decline in neuronal populations might contribute to the loss of 286 motor skills and cognitive function in the mice which we saw during the psychological and 287 rotarod experiments (Figure 6 and Figure 7). 288 The hippocampus functions in memory and learning and GABBR2 expression 289 typically occurs in the area near mossy fiber synapses that form the major excitatory input 290 into the auto-associative network of pyramidal cells in the CA3 region (34). The loss of 291 inhibitory input by GABBR2 to CA3 neurons could produce excitotoxity resulting of loss of 292 pyramidal neurons (Figure 4B). Loss of neurons that govern lateral inhibition in the 293 dentate gyrus can result in delamination of the granule cell layer and multilamellar 294 discharges in response to cortical stimuli resulting in increased excitotoxicity (35). In the 295 GABBR2 cKO mouse, progressive damage over time to the excitable neurons lacking 296 GABBR2 input in the dentate gyrus of the hippocampus likely results in susceptibility to 297 seizures and hyperexcitability (Figure 6). The histological phenotype of the GABBR2 cKO 298 hippocampus is reminiscent of patients with epilepsy with a loss of dentate hilar neurons 299 that govern dentate granule cell excitability (36, 37). 300 Cerebellar Purkinje neurons are known to express GABBR2 (38). Purkinje neurons 301 project to the intermediate discharge layer and are the key efferent output of the 302 cerebellum. The GABBR2 cKO mice have fewer Purkinje neurons (Figure 5, yellow arrows) 303 and smaller dendritic arbors (Figure 5, red arrows) contributing to the abnormal rotarod 304 performance. Purkinje dysfunction may also lead to fewer connections between the 305 Purkinje neuron’s dendritic arbors and the interneurons in the molecular layer which may 306 increase glutamatergic stimuli and neurotoxicity. Some seizure disorders cause increased 307 Purkinje death by glutamate excitotoxicity (39). .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 15 308 In conclusion, the phenotype of GABBR2 cKO mice was similar to the global GABBR2 309 KO mouse (3). Our characterization of the Actin-Cre GABBR2 cKO mice has revealed 310 histological changes in multiple areas of the brain in addition to changes in rearing 311 behavior and premature death. Our studies do not discriminate between whether these 312 changes are the result of altered brain development or due to progressive excitotoxic 313 neuronal damage, although the use of inducible Cre transgenes could address this question 314 in the future. Nevertheless, these studies demonstrate that the floxed mice reported here 315 will provide a new tool to target tissue-specific GABBR2 signaling through Cre-mediated 316 recombination. This will now provide scientists the ability to study GABBR2 function in 317 different cell types without the potentially confounding effects of the neurological 318 dysfunction caused by global knockout of GABBR2. 319 320 Acknowledgments: 321 We thank the Yale Genome Editing Center for their help in generating the GABBR2 cKO 322 mouse. 323 324 325 326 Figure 1 327 CRISPR design to add loxP sites to the Gabbr2 gene. A. Map of Gabbr2 gene showing the 328 guide RNAs used to create LoxP sites. B. Primers used to identify loxP sites in Actin-CRE 329 Gabbr2lox/lox mice. C. Table summarizing the different Gabbr2lox/lox mouse lines generated. 330 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 16 331 Figure 2. 332 Expression of Gabbr2, Gabbr1, and Casr in the brains of control and cKO mice. A. QPCR to 333 assess mRNA levels in whole-brain RNA. The specific transcript is shown on the top of each 334 graph. Bars represent the mean ± SEM. *** p<.001. B. Protein expression levels of 335 GABBR2, GABBR1, and CASR in whole-brain extracts. n= 6 control and n=12 cKO. 336 337 Figure 3. 338 Brain sections of cerebral cortex from cKO and control brains stained for S100 and Luxol 339 blue. There was a reduction of S100 staining overall and fewer S100-labeled neurons in 340 the cKO cortex (A) versus control cortex (B). Yellow arrows point to neuronal dendrites. 341 Red arrows point to S100-labeled neuronal bodies. There are more vacuolated neurons 342 and cell debris in the cortex of cKO mice revealed by Luxol blue staining (C versus D). 343 Green arrows point examples of vacuolated neurons. Scale bar = 200 microns 344 345 Figure 4. 346 Hippocampus of the cKO mice have increased vacuolation of pyramidal neurons and a 347 reduction of CA3 neurons and granular cells when compared to control mice. The 348 polymorphic layer of dentate gyrus (A and B) with CA3 neurons are identified with blue 349 arrows, the dense immature granules with green arrows. The white dotted line showing 350 border of the dense immature granular layer, and yellow arrows identify pyramidal 351 neurons of polymorphic layer, and red arrows identify granular cells. A congested capillary 352 is present in the cKO hippocampus (black arrow). Hematoxylin and eosin staining of the .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 17 353 hippocampus in A-B, and E -F is Luxol Blue staining. A and C are control mice, B and D are 354 cKO mice. Scale bar = 1000 microns in A and B, 200 microns in the insets, and 100 microns 355 in C and D. 356 357 Figure 5. 358 Cerebellum of cKO and control mice stained with Luxol blue. A is a control brain and B is a 359 representative cKO brain. The cerebellum has reduced Luxol blue staining in the cKO 360 compared to the control brain. There are fewer Purkinje neurons (yellow arrows) and 361 shorter, fewer and less complex dendritic extensions (red arrows) in the cKO cerebellum 362 (A versus B). A and B Scale bar is 1000 microns. Inserts of A and B scale bar is 200 microns. 363 364 Figure 6. 365 Measurements of locomotor activity and unsupported rearing in cKO versus control mice. 366 Increased locomotor activity and decreased unsupported rearing is seen in cKO mice. Bars 367 represent the mean ± SEM. ***p<0.001, **p<0.01. n= 14 for control mice, n=11 for cKO 368 mice. 369 370 Figure 7. 371 Rotarod experiments in female and male control and cKO mice. A). Training periods over 2 372 separate days showing times on rotarod for control (blue) and cKO mice (red). Points 373 represent the mean ± SEM dwelling times over 5 trials on each of 2 days. Note that cKO 374 mice stay on the rod for shorter periods at baseline, although they do improve with 375 training. B). Rotarod experiments in female and male control and cKO mice before and .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 18 376 after administration of baclofen, a GABBR2 agonist. Bar graphs on the left demonstrate the 377 absolute dwell times on the rotarod. Bar graphs on the right represent the change in dwell 378 time in response to baclofen administration. Bars represent mean ± SEM. *P<0.05, 379 **p<0.01, ***p<0.001. n=4/group 380 381 Figure 8. 382 Kaplan-Myer plot showing survival curves of control and cKO mice. No control mice died 383 over 150 days of observation. 100% of cKO mice died by 115 days. n= 11 in each group. 384 Gehan-Breslow-Wilcoxon test (p<0.0001). 385 386 387 388 389 References: 390 1. Misgeld U, Bijak M, Jarolimek W. A physiological role for GABAB receptors and the effects of 391 baclofen in the mammalian central nervous system. Prog Neurobiol. 1995;46(4):423-62. 392 2. Couve A, Moss SJ, Pangalos MN. GABAB receptors: a new paradigm in G protein signaling. Mol 393 Cell Neurosci. 2000;16(4):296-312. 394 3. Gassmann M, Shaban H, Vigot R, Sansig G, Haller C, Barbieri S, et al. Redistribution of GABAB(1) 395 protein and atypical GABAB responses in GABAB(2)-deficient mice. J Neurosci. 2004;24(27):6086-97. 396 4. Cheng Z, Tu C, Rodriguez L, Chen TH, Dvorak MM, Margeta M, et al. Type B gamma- 397 aminobutyric acid receptors modulate the function of the extracellular Ca2+-sensing receptor and cell 398 differentiation in murine growth plate chondrocytes. Endocrinology. 2007;148(10):4984-92. 399 5. Zhang D, Li X, Yao Z, Wei C, Ning N, Li J. GABAergic signaling facilitates breast cancer metastasis 400 by promoting ERK1/2-dependent phosphorylation. Cancer Lett. 2014;348(1-2):100-8. 401 6. Wu JX, Shan FX, Zheng JN, Pei DS. beta-arrestin promotes c-Jun N-terminal kinase mediated 402 apoptosis via a GABA(B)R.beta-arrestin.JNK signaling module. Asian Pac J Cancer Prev. 2014;15(2):1041- 403 6. 404 7. Marshall FH, White J, Main M, Green A, Wise A. GABA(B) receptors function as heterodimers. 405 Biochem Soc Trans. 1999;27(4):530-5. 406 8. Mohler H, Benke D, Fritschy JM. GABA(B)-receptor isoforms molecular architecture and 407 distribution. Life Sci. 2001;68(19-20):2297-300. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 19 408 9. Chang W, Tu C, Cheng Z, Rodriguez L, Chen TH, Gassmann M, et al. Complex formation with the 409 Type B gamma-aminobutyric acid receptor affects the expression and signal transduction of the 410 extracellular calcium-sensing receptor. Studies with HEK-293 cells and neurons. J Biol Chem. 411 2007;282(34):25030-40. 412 10. Chang W, Tu CL, Jean-Alphonse FG, Herberger A, Cheng Z, Hwong J, et al. PTH hypersecretion 413 triggered by a GABA(B1) and Ca(2+)-sensing receptor heterocomplex in hyperparathyroidism. Nat 414 Metab. 2020;2(3):243-55. 415 11. Yang H, Wang H, Jaenisch R. Generating genetically modified mice using CRISPR/Cas-mediated 416 genome engineering. Nat Protoc. 2014;9(8):1956-68. 417 12. Price NL, Rotllan N, Zhang X, Canfran-Duque A, Nottoli T, Suarez Y, et al. Specific Disruption of 418 Abca1 Targeting Largely Mimics the Effects of miR-33 Knockout on Macrophage Cholesterol Efflux and 419 Atherosclerotic Plaque Development. Circ Res. 2019;124(6):874-80. 420 13. Quadros RM, Miura H, Harms DW, Akatsuka H, Sato T, Aida T, et al. Easi-CRISPR: a robust 421 method for one-step generation of mice carrying conditional and insertion alleles using long ssDNA 422 donors and CRISPR ribonucleoproteins. Genome Biol. 2017;18(1):92. 423 14. Concordet JP, Haeussler M. CRISPOR: intuitive guide selection for CRISPR/Cas9 genome editing 424 experiments and screens. Nucleic Acids Res. 2018;46(W1):W242-W5. 425 15. Nagy A. Manipulating the mouse embryo : a laboratory manual. 3rd ed. Cold Spring Harbor, N.Y.: 426 Cold Spring Harbor Laboratory Press Cold Spring Harbor, N.Y.; 2003. 427 16. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative 428 PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4):402-8. 429 17. Carson FL, Cappellano CH. Histotechnology. A self-instructional text. 3rd ed. [Chicago]: ASCP 430 Press [Chicago]; 2009. 431 18. Schuler V, Luscher C, Blanchet C, Klix N, Sansig G, Klebs K, et al. Epilepsy, hyperalgesia, impaired 432 memory, and loss of pre- and postsynaptic GABA(B) responses in mice lacking GABA(B(1)). Neuron. 433 2001;31(1):47-58. 434 19. During MJ, Ryder KM, Spencer DD. Hippocampal GABA transporter function in temporal-lobe 435 epilepsy. Nature. 1995;376(6536):174-7. 436 20. Mederos S, Sanchez-Puelles C, Esparza J, Valero M, Ponomarenko A, Perea G. GABAergic 437 signaling to astrocytes in the prefrontal cortex sustains goal-directed behaviors. Nat Neurosci. 438 2021;24(1):82-92. 439 21. Scanziani M. GABA spillover activates postsynaptic GABA(B) receptors to control rhythmic 440 hippocampal activity. Neuron. 2000;25(3):673-81. 441 22. Wang L, Bruce G, Spary E, Deuchars J, Deuchars SA. GABA(B) Mediated Regulation of 442 Sympathetic Preganglionic Neurons: Pre- and Postsynaptic Sites of Action. Front Neurol. 2010;1:142. 443 23. Kuner R, Kohr G, Grunewald S, Eisenhardt G, Bach A, Kornau HC. Role of heteromer formation in 444 GABAB receptor function. Science. 1999;283(5398):74-7. 445 24. Chong ZZ, Changyaleket B, Xu H, Dull RO, Schwartz DE. Identifying S100B as a Biomarker and a 446 Therapeutic Target For Brain Injury and Multiple Diseases. Curr Med Chem. 2016;23(15):1571-96. 447 25. Hernandez-Ortega K, Canul-Euan AA, Solis-Paredes JM, Borboa-Olivares H, Reyes-Munoz E, 448 Estrada-Gutierrez G, et al. S100B actions on glial and neuronal cells in the developing brain: an overview. 449 Front Neurosci. 2024;18:1425525. 450 26. Mombereau C, Kaupmann K, Froestl W, Sansig G, van der Putten H, Cryan JF. Genetic and 451 pharmacological evidence of a role for GABA(B) receptors in the modulation of anxiety- and 452 antidepressant-like behavior. Neuropsychopharmacology. 2004;29(6):1050-62. 453 27. Lever C, Burton S, O'Keefe J. Rearing on hind legs, environmental novelty, and the hippocampal 454 formation. Rev Neurosci. 2006;17(1-2):111-33. .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint 20 455 28. Torres-Lista V, Parrado-Fernandez C, Alvarez-Monton I, Frontinan-Rubio J, Duran-Prado M, 456 Peinado JR, et al. Neophobia, NQO1 and SIRT1 as premorbid and prodromal indicators of AD in 3xTg-AD 457 mice. Behav Brain Res. 2014;271:140-6. 458 29. Sturman O, Germain PL, Bohacek J. Exploratory rearing: a context- and stress-sensitive behavior 459 recorded in the open-field test. Stress. 2018;21(5):443-52. 460 30. Davidoff RA. Antispasticity drugs: mechanisms of action. Ann Neurol. 1985;17(2):107-16. 461 31. Allerton CA, Boden PR, Hill RG. Actions of the GABAB agonist, (-)-baclofen, on neurones in deep 462 dorsal horn of the rat spinal cord in vitro. Br J Pharmacol. 1989;96(1):29-38. 463 32. Kelsom C, Lu W. Development and specification of GABAergic cortical interneurons. Cell Biosci. 464 2013;3(1):19. 465 33. Zeng X, Niu Y, Qin G, Zhang D, Zhou J, Chen L. Deficiency in the function of inhibitory 466 interneurons contributes to glutamate-associated central sensitization through GABABR2-SynCAM1 467 signaling in chronic migraine rats. FASEB J. 2020;34(11):14780-98. 468 34. Vogt KE, Nicoll RA. Glutamate and gamma-aminobutyric acid mediate a heterosynaptic 469 depression at mossy fiber synapses in the hippocampus. Proc Natl Acad Sci U S A. 1999;96(3):1118-22. 470 35. Sloviter RS. The functional organization of the hippocampal dentate gyrus and its relevance to 471 the pathogenesis of temporal lobe epilepsy. Ann Neurol. 1994;35(6):640-54. 472 36. Cendes F. Febrile seizures and mesial temporal sclerosis. Curr Opin Neurol. 2004;17(2):161-4. 473 37. Yang F, Liu ZR, Chen J, Zhang SJ, Quan QY, Huang YG, et al. Roles of astrocytes and microglia in 474 seizure-induced aberrant neurogenesis in the hippocampus of adult rats. J Neurosci Res. 475 2010;88(3):519-29. 476 38. Jamal L, Khan AN, Butt S, Patel CR, Zhang H. The level and distribution of the GABA(B)R1 and 477 GABA(B)R2 receptor subunits in the rat's inferior colliculus. Front Neural Circuits. 2012;6:92. 478 39. Paul MS, Limaiem F. Histology, Purkinje Cells. StatPearls. Treasure Island (FL)2024. 479 480 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 23, 2025. ; https://doi.org/10.1101/2025.01.23.634473doi: bioRxiv preprint