Conditional ablation of vasopressin‐synthesizing neurons in transgenic rats

Abstract Vasopressin‐synthesizing neurons are located in several brain regions, including the hypothalamic paraventricular nucleus (PVN), supraoptic nucleus (SON) and suprachiasmatic nucleus (SCN). Vasopressin has been shown to have various functions in the brain, including social recognition memory, stress responses, emotional behaviors and circadian rhythms. The precise physiological functions of vasopressin‐synthesizing neurons in specific brain regions remain to be clarified. Conditional ablation of local vasopressin‐synthesizing neurons may be a useful tool for investigation of the functions of vasopressin neurons in the regions. In the present study, we characterized a transgenic rat line that expresses a mutated human diphtheria toxin receptor under control of the vasopressin gene promoter. Under a condition of salt loading, which activates the vasopressin gene in the hypothalamic PVN and SON, transgenic rats were i.c.v. injected with diphtheria toxin. Intracerebroventricular administration of diphtheria toxin after salt loading depleted vasopressin‐immunoreactive cells in the hypothalamic PVN and SON, but not in the SCN. The number of oxytocin‐immunoreactive cells in the hypothalamus was not significantly changed. The rats that received i.c.v. diphtheria toxin after salt loading showed polydipsia and polyuria, which were rescued by peripheral administration of 1‐deamino‐8‐d‐arginine vasopressin via an osmotic mini‐pump. Intrahypothalamic administration of diphtheria toxin in transgenic rats under a normal hydration condition reduced the number of vasopressin‐immunoreactive neurons, but not the number of oxytocin‐immunoreactive neurons. The transgenic rat model can be used for selective ablation of vasopressin‐synthesizing neurons and may be useful for clarifying roles of vasopressin neurons at least in the hypothalamic PVN and SON in the rat.


| INTRODUC TI ON
Vasopressin, an evolutionally conserved peptide consisting of nine amino acids, shows similarities throughout the animal kingdom. [1][2][3] Vasopressin not only has anti-diuretic and vasopressor actions, but also affects social behaviors. 4 The majority of vasopressin neurons are located within the hypothalamus including the hypothalamic paraventricular nucleus (PVN), supraoptic nucleus (SON) and suprachiasmatic nucleus (SCN). Vasopressin neurons in each brain region project to different target regions of the brain and are considered to exert different functions. 5 To clarify region-specific functions of vasopressin neurons, region-specific experimental manipulation is considered to be effective. The human diphtheria toxin receptor shows more than 1000-fold higher sensitivity to a diphtheria toxin than the receptor of rodents does. 6 Selective induction of the human diphtheria toxin receptor in a certain type of cells of mice has been used for selective destruction of cells in a time-dependent manner. 7 Here, we report transgenic rats for which vasopressin neurons selectively express the human diphtheria toxin receptor. These rats were used to clarify the function of vasopressin neurons in the olfactory bulb. 8 However, the detailed characteristics of these rats remain to be examined. We examined the expression of the diphtheria toxin receptor in the hypothalamus and investigated the effects of i.c.v. and intrahypothalamic injections of diphtheria toxin on the numbers of vasopressin neurons and oxytocin neurons. Lastly, the effects of destruction of hypothalamic PVN vasopressin neurons on social recognition, on object recognition memory and on social interaction were examined using the transgenic animal models because vasopressin receptors in the hippocampus have been shown to be involved in social recognition memory and, in addition, vasopressin neurons in the hypothalamic PVN project to the hippocampus. 9

| Animals
Male Lewis rats (LEW/CrlCrlj) were obtained from Charles River Laboratories Japan, Inc. Transgenic rats expressing human heparinbinding epidermal growth factor-like growth factor (hHB-EGF) as a diphtheria toxin receptor under control of the vasopressin promoter were created as described below. Male rats were used for all experiments. Rats were housed under a 12:12 h light/dark photocycle (lights on 7.30 pm) at 22 ± 2°C and 40%-70% relative humidity with food and water available ad lib. Rats were group-housed before the surgical operation. After surgery, the rats were kept in individual cages. All animal procedures were approved by the Judging Committee of Experimental Animal Ethics of Jichi Medical University and were conducted in accordance with Japanese legislation concerning animal experiments.

| Generation of transgenic rats
The plasmids p5-VCAT-3-Sal and prVP8.2R1 10  was microinjected as a transgene into pronuclei of fertilized one-cell embryos from Lewis rats (LEW/CrlCrlj; Charles River Laboratories Japan). These microinjected embryos were then transferred into pseudopregnant surrogate female rats. The transgenic founders were then mated with Lewis rats.

| Reverse transcriptase-polymerase chain reaction (RT-PCR)
Brains were isolated immediately after decapitation of 7-8-month-old transgenic and wild-type rats, frozen on dry ice and stored at −80°C.
The hypothalamic part of the frozen brains was sectioned coronally at 300 μm in a cryostat at −10°C and then thaw-mounted onto glass slides and refrozen. The regions containing the SON were micropunched from these sections that were kept frozen on dry ice using a 0.5-mm-diameter Sample Corer (Fine Science Tools) under a microscope. The regions containing the hypothalamic PVN and suprachiasmatic nuclei were micro-punched from these sections kept frozen on dry ice using a 1. to detect hHB-EGF and β-actin mRNAs as a positive control for the presence of cDNA in samples and for PCR amplification. Amplified RT-PCR products were visualized by electrophoresis on 2% agarose gels stained with ethidium bromide.
The expected sizes of PCR products for hHB-EGF and β-actin mRNA were 177 and 150 bp, respectively.

| Droplet digital PCR
A QX200 droplet digital PCR system (Bio-Rad Laboratories, Inc.) was used for determination of transgene copy number. Genomic DNA was digested with restriction enzymes, MseI and SmaI. Rat F I G U R E 1 Generation of vasopressin-human heparin-binding epidermal growth factor-like growth factor (hHB-EGF) transgenic rats. (A) Structure of the transgene construct. The construct consists of the vasopressin structural gene, hHB-EGF cDNA, rabbit β-globin polyadenylation (poly A) signal and simian virus 40 (SV40) early gene polyadenylation signal. E1-E3, exon 1-exon 3. Restriction enzyme sites are shown below the transgene construct. Ec, EcoRI; N, NotI; S, SalI; SII, SacII. (B) Analysis of hHB-EGF mRNA expression in the hypothalamic supraoptic nucleus (SON), hypothalamic paraventricular nucleus (PVN) and suprachiasmatic nucleus (SCN) of transgenic (Tg (+)) and wild-type (Tg (-)) rats by a reverse transcriptase-polymerase chain reaction (RT-PCR). The micropunched SON, PVN and SCN were evaluated for expression levels of hHB-EGF mRNA by a RT-PCR. Beta-actin (Actb), a house-keeping gene, was used as a positive control. (C) Immunostaining of the SON of salt-loaded transgenic rats with anti-vasopressin antibody (AVP; magenta) and anti-hHB-EGF antibody (hHB-EGF; green). (D) Percentage of vasopressin-immunoreactive cells expressing hHB-EGF in the SON of salt-loaded transgenic rats. **p < .01 vs. salt-loaded wild-type rats. Scale bar = 20 μm oxytocin receptor gene was used as a reference gene for normalization of hHB-EGF copy number. The assay was performed in a 20-μL reaction volume containing ddPCR supermix (Bio-Rad Laboratories, Inc.), hydrolysis probes and gene-specific primers. The hydrolysis probes and PCR primers sets that were used were: rat

| Urine collection and measurement of urine osmolality and vasopressin concentrations
Rats i.c.v. injected with diphtheria toxin were housed in metabolic cages (3701M081; Tecniplast) for 6-8 days. After habituation in the cages for more than 1 day, urine was collected once per day and the volume was measured. Urine osmolality was measured by freezing point depression with an osmometer (Fiske Micro-Osmometer Model 210; Fiske Associates) as described previously. 13 Volume of water intake, volume of urine, urine osmolality and vasopressin concentrations were shown as an average of 3 consecutive days. Vasopressin concentrations were determined by radioimmunoassay with specific anti-arginine vasopressin antibody, as described previously. 14

| Surgical operation for injections of diphtheria toxin
For i.c.v. injection of diphtheria toxin, a surgical operation of i.c.v.
cannula implantation was performed in transgenic rats and wildtype rats as described previously. 15  with diphtheria toxin (10 ng per 5 μL, once per day) on 3 consecutive days of days 3-5 after initiation of salt loading and were implanted with osmotic pumps s.c. for administration of a vasopressin V2 receptor agonist as described below.
For intrahypothalamic injection of diphtheria toxin under a normal osmotic condition, transgenic rats and wild-type rats at the age of 5 months were used. Transgenic rats and wild-type rats  rats out of 13 rats of the group injected with diphtheria toxin was more than half that of control rats and so the behavioral data were analyzed excluding these two rats.
The brains were removed, postfixed in 4% PFA overnight and transferred to 30% sucrose solution in 0.1 m phosphate buffer until tissues sank. The brains were frozen on dry ice and stored at -80°C. For detection of vasopressin and oxytocin-immunoreactive cells, brain sections were prepared as described above and processed as described previously. 16 In brief, sections were incubated with a guinea pig polyclonal antibody against vasopressin (dilution  . Their behaviors were video-monitored for 2 days. Images from each cage were captured at a rate of one frame per second. Social interaction was measured by counting the number of particles in each frame: two particles indicated that the rats were not in contact with each other and one particle indicated contact between the two rats.

| Statistical analysis
All statistical analyses were performed using Prism (GraphPad Software Inc.

| Generation of transgenic rats expressing hHB-EGF under control of the vasopressin promoter
The sensitivity towards diphtheria toxin varies among mammalian species. Humans show high sensitivity, whereas rats have resistance to diphtheria toxin. For conditional ablation of vasopressin neurons, we created transgenic rats that expressed a mutated hHB-EGF gene under control of the vasopressin promoter ( Figure 1A).
The mutated hHB-EGF has a high affinity to diphtheria toxin, but little EGF-like activity. 12 Transgene copy number in these transgenic rats determined by droplet digital PCR was found to be 8-9 copies of transgene per haploid genome (8.61 ± 0.05, n = 13). The mutated hHB-EGF mRNA was detected in the SON, hypothalamic PVN and suprachiasmatic nuclei of transgenic rats, in which vasopressin neurons were located ( Figure 1B). Immunoreactivity for hHB-EGF was also examined ( Figure 1C). The transgene of the mutated hHB-EGF was inserted into the downstream region of the vasopressin promoter, which is activated by salt-loading.
Immunoreactivity for hHB-EGF was examined in rats under a normal osmotic condition and in rats that received 2% NaCl as drinking water for 1 week. In salt-loaded transgenic rats, the percentage of vasopressin-immunoreactive neurons expressing hHB-EGF was significantly higher than the percentages in non-transgenic wildtype rats and non-salt-loaded transgenic rats ( Figure 1D,

| I.c.v. injection of diphtheria toxin in transgenic rats
We then investigated whether i.c.v. injection of diphtheria toxin Student's t test).

| I.c.v. injection of diphtheria toxin in salt-loaded transgenic rats
To upregulate the expression of the transgene and to induce efficient ablation of vasopressin neurons, salt loading was applied in transgenic rats and in wild-type rats before diphtheria toxin injection. Salt loading was applied by 2% NaCl drinking water. Diphtheria toxin was injected i.c.v. once a day for 3 days and daily water intake was measured ( Figure 3A). Amounts of water intake and urine excretion, urine osmolarity and urine vasopressin concentrations t test) ( Figure 3D). In addition, the amount of water intake was increased again to 69.6 ± 18.7 mL day -1 in transgenic rats 65 days after implantation of DDAVP osmotic mini-pumps as a result of depletion of DDAVP in the implanted pumps. From these results, phenotypes observed following i.c.v. injections with diphtheria toxin were considered to be caused by central diabetes insipidus.

| Number of vasopressin neurons after i.c.v. injection of diphtheria toxin
To examine the specificity of destruction of vasopressin neurons in transgenic rats injected with diphtheria toxin, vasopressin immunoreactivity and oxytocin immunoreactivity in the brains of these rats were determined by immunohistochemistry.

| Number of vasopressin neurons following intrahypothalamic injection of diphtheria toxin
To examine whether injections of diphtheria toxin into brain parenchyma under a normal osmotic condition destroy vasopressin

| Social behaviors in transgenic rats with destruction of vasopressin neurons in the PVN
Vasopressin in the hippocampus has been shown to play an important role in social memory. 9 Infusion of antiserum for vasopressin into the hippocampus impairs social memory. 19 The hippocampus Student's t test) ( Figure 7B).
In two rats out of 13 rats of the diphtheria toxin-injected group, the number of vasopressin neurons was more than half compared to that in vehicle-injected rats. To determine the effects of ablation of PVN vasopressin neurons on behaviors, behavioral data were analyzed excluding these two rats. In a social recognition test, the time spent for sniffing a stimulus rat in the training session and preference index for a novel stimulus rat in the test session were not significantly different between vehicle-injected rats and vasopressin neuron-ablated rats (sniffing time: U = 44, p = .0726; preference index: U = 64, p = .5007; Mann-Whitney U test) ( Figure 7C).
We also examined object recognition memory as non-social memory. In an object recognition test, the time spent for sniffing a stimulus object in the training session and preference index for a novel object in the test session were not significantly different between vehicle-injected rats and vasopressin neuron-ablated rats (sniffing time: U = 46, p = .0954; preference index: U = 75, p = .9358; Mann-Whitney U test) ( Figure 7D).
Vasopressin V1a receptor-deficient mice have been reported to exhibit deficits in social interaction. 21 Thus, we also examined social interaction in these rats. In a home cage social interaction test, did not find significant differences in social behaviors in these transgenic rats.

| D ISCUSS I ON
The human diphtheria toxin receptor, hHB-EGF, has 1000-fold higher sensitivity to diphtheria toxin compared to that of rodents, and selective induction of the receptor using a specific promoter has been employed for selective destruction of specific cells. 7 Here, we char-

CO N FLI C T O F I NTE R E S T S
The authors declare that they have no conflicts of interest.

DATA AVA I L A B I L I T Y
The data that support the findings of this study are available from the corresponding author upon reasonable request.