Transcriptomic Alterations in the Hippocampus and Prefrontal Cortex of Rats with Chronic Unpredictable Stress Induced by Low-Intensity Pulsed Ultrasound

Abstract

In our previous study, low-intensity pulsed ultrasound stimulation (LIPUS), a novel noninvasive neuromodulation tool, effectively alleviated depression-like behaviors in a rat model of depression. However, the underlying mechanisms remain unclear. The present study aimed to elucidate the relevant mechanisms through transcriptomic research. To induce depression-like phenotypes, rats were subjected to chronic unpredictable stress (CUS) for six weeks. Following this initial stress period, LIPUS was administered daily for an additional four weeks while the rats were continuously exposed to CUS, which was previously determined to significantly reduce immobility time in forced swimming test. The transcriptome expression profiles in the hippocampus and prefrontal cortex (PFC) were subsequently analyzed by RNA sequencing techniques. Overall, LIPUS reversed the abnormal expression of genes in the brain of model rats, especially the expression of genes linked to glucose and lipid metabolism. Specifically, we found that the CUS rats exhibited 3464 differentially expressed genes (DEGs) in the hippocampus and 1781 DEGs in the PFC compared to the control group. LIPUS reversed the expression of 592 genes in the hippocampus and 254 genes in the PFC. Functional analysis revealed a significant enrichment of DEGs related to oxygen carrier activity and sugar metabolism. Subsequently, the alterations in the top ten genes were validated using real-time PCR. The significant changes in the expression of hemoglobin subunit beta, growth hormone 1, and glucokinase were confirmed. The results suggest that LIPUS may alleviate depression-related manifestations by regulating cerebral oxygenation and sugar metabolism processes in the brain.

Graphical Abstract

Introduction

According to the China Mental Health Survey, the weighted lifetime prevalence of depressive disorders is 8.0% in females and 5.7% in males [1]. Although the precise etiology of depression remains unclear, it is widely recognized that a combination of environmental and genetic factors contribute to its development. Recent studies have highlighted the role of environmental factors, such as stress, in gene expression through epigenetic mechanisms including DNA methylation, histone modification, and micro-RNA expression, thereby playing a significant role in the pathogenesis of depression [2, 3].

Antidepressants are the most commonly used treatment in clinical practice and have demonstrated significant improvement in depressive disorders. However, their delayed onset of action and the occurrence of various unpleasant side effects, such as headache, insomnia, nausea, and weight changes, have limited their utility [4, 5]. For example, the time needed for the onset of action for first-line pharmaceuticals was typically 4–6 weeks or even longer [6]. Additionally, more than 30% of patients do not respond to antidepressants, placing a substantial burden on families and society. Therefore, we need to continue the search for alternative effective therapies.

Neuromodulation therapy serves as an adjunctive therapeutic approach for neuropsychiatric disorders. Among various neuromodulation techniques, ultrasound stimulation has emerged as a promising strategy for treating central nervous system diseases due to its high efficiency, non-invasiveness, and favorable side effects [7, 8]. Furthermore, accumulating evidence supports the efficacy of ultrasound stimulation as therapeutic intervention for depression. For instance, a study by Kim et al. demonstrated a reduction in depression and anxiety symptoms in patients with refractory obsessive–compulsive disorder following ultrasound stimulation [9]. Additionally, animal experiments have demonstrated that ultrasound stimulation can significantly improve depressive-like behaviors in the rat model of depression, evidenced by an increased sucrose preference index in the sucrose preference test and reduced immobility time in the forced swimming test (FST) [10].

In our previous studies, we found that low-intensity pulsed ultrasound stimulation (LIPUS) effectively alleviated depressive symptoms in rats subjected to chronic unpredictable stress (CUS) [11]. Importantly, we observed no indication of micro-hemorrhage, edema, necrosis, or inflammation following LIPUS. Although we provided evidence linking the antidepressant effects of LIPUS to the mammalian target of rapamycin (mTOR) signaling pathway, the limited number of studies on the underlying mechanisms hampers the clinical utility of LIPUS.

Therefore, the objective of this study was to investigate the changes in the transcriptome profile induced by LIPUS and explore its molecular mechanisms. We aimed to assess RNA expression in the hippocampus and prefrontal cortex (PFC), validate the differentially expressed genes through quantitative PCR (qPCR) analysis, and elucidate the effect of LIPUS on gene expression. Finally, we sought to identify the pathways involved in the effects of LIPUS.

Materials and Methods

Animal Experiment

The samples utilized in this experiment comprised the remaining right hippocampus and PFC obtained from animals employed in a previously published study [11]. In our previous study, we assessed the behavioral effects of single stimulation with LIPUS at three different pulse frequencies (200 Hz, 285 Hz and 500 Hz) for 30 min on vmPFC in non-stressed rats. Only 200 Hz LIPUS significantly reduced immobilization time in the FST and increased the number of c-Fos-positive cells in the ventromedial PFC. Therefore, we used the frequency of 200 HZ in subsequent experiments. The experimental animal conditions, the method for establishing the depression model through chronic unpredictable stress, the LIPUS parameters, and the sample collection methods were consistent with the details outlined in a previous publication. The experimental schedule was as follows: after 1 week of adaptation, rats in the control group were fed normally in pairs. Rats in the CUS group were housed in individual cages and were subjected to one or two of fourteen randomly selected different stressors per day for six consecutive weeks. After completing the CUS exposure, each rat was fixed with an ultrasound collimator on the skull and was given 1 week to recover from surgery. Normal control rats underwent the same craniotomy procedure, but the collimator was not affixed. Next, ultrasound stimulation was employed to awake and freely move rats in the CUS + LIPUS group, for twenty minutes each day for four consecutive weeks. Rats in the control and CUS + sham LIPUS groups underwent the same procedures, but the power was turned off. Behavioral tests were conducted immediately after LIPUS. Thereafter, rats were sacrificed by intraperitoneal injection of pentobarbital sodium 150 mg/kg, and their brains were excised for morphological and biochemical experiments (Fig. 1A).

Fig. 1.

LIPUS induced antidepressant-like effects. A shows the experimental schedule of CUS and LIPUS treatment, and behavior tests. B–C Latency time of exploring and eating in novelty-suppressed feeding test. D–E Latency time and number of shuttles crossing in light/dark test. Data are expressed as mean ± SEM (n = 8 per group). *p < 0.05 vs. the control group; #p < 0.05 vs. the CUS group

Experimental Animals

In total, thirty 6–8-week-old male Sprague–Dawley rats (150–180 g), provided by Beijing Vital River Laboratory Animal Technology Co., Ltd. were housed at a controlled ambient temperature of 23 ± 2 ℃, 50 ± 10% relative humidity under a 12-h light–dark cycle. They had free access to food and water. The study protocol was approved by the Animal Use and Care Committee at Capital Medical University (AEEI-2018–003). All efforts were made to minimize animal suffering.

CUS Procedures

All rats were randomly divided into the control (n = 8) or CUS groups (n = 22) based on the sucrose performance index and housed separately. Rats in the control group rats were pair-housed in cages with ad libitum access to food and water, except during performance tests. Rats in the CUS group rats were individually housed and subjected to 1–2 stressors per day over six weeks. The stressors included 36 V electric foot-shock, forced swimming in cold water (4 ℃) or hot water (45 ℃), immobilization, water and food deprivation, tail clamp, wet bedding, no padding, soiled cage, cage titling at 45°, 85 dB noise, continuous illumination and stroboscopic illumination. At the end of the 6-week CUS procedure, 16 rats successfully underwent the depression models, defined by significantly reduced body weight and sucrose preference indices (SPI) compared to control group rats. The 16 model rats were redistributed into CUS and LIPUS groups based on the SPI, which was calculated as follows: sucrose consumption (g)/total fluid consumption (g) × 100%.

Low-Intensity Pulsed Ultrasound

Rats in the LIPUS group underwent surgical procedures to affix a collimator with drilled holes in the skull. LIPUS was transduced using a homemade wearable transducer (frequency: 800 kHz), made by 1–3 PZT-5H/epoxy composite, epoxy backing, and Parylene layer. A polymethyl methacrylate collimator was employed to fix the transducer onto the target region of the skull. The total length of the collimator was approximately 9 mm from the face of the transducer to the end of the collimator, with an upper internal diameter of 14 mm. At 4 mm from the face of the transducer, there was a step in the collimator, and the diameter at the end of the collimator was 8 mm. During ultrasound stimulation, the transducer was connected to the collimator. The collimator was filled with deionized, degassed water, to ensure successful delivery of ultrasound energy into the brain. The parameters of LIPUS were applied with 200 Hz pulse repetition frequency, 4% duty cycle, 0.2 ms tone burst duration, 1-s sonication duration, and 3-s inter-stimulus interval.

Behavioral Tests

The sucrose preference test was done at the beginning of the experiment and at the end of each week. The open field test (OFT), FST, novelty suppressed feeding test, and light–dark transition test were conducted at the end of the fourth week of LIPUS. The sucrose preference test was used to assess anhedonia, the open field test was conducted to evaluate the locomotor ability of mice after surgery, and the forced swim test was employed to model despair behavior associated with depression. These behavioral outcomes have been previously reported in our published articles [11]. In this manuscript, we introduce two additional behavioral experiments to assess depressive-like behaviors.

In the novelty suppressed feeding test (NSFT), the animals were fasted for 24 h before the experiment. An open field (100 × 100 × 40 cm) was used in the novelty suppressed feeding test. Blocks of rat chow were place in the center. The rat was gently placed, facing outward, in the corner of the box. The whole procedure was video-recorded. The latency time before rats’ head goes into the middle cell and it begins to bite the food reflected the degree of anxiety/depression.

Light–dark transition test (LDT) was based on the innate aversion of rodents to the brightly illuminated and open areas. Dark box had the top cover and the light box was open and illuminated. There was a hole between the two boxes for animals to pass through. The rats were gently put in the light box and allowed to move freely in the dark and light box. The time of when the animals completely entered the dark room from the light room for the first time and the number of times shuttling through the partition hole within 5 min was recorded.

Clustering and Sequencing

For index-coded sample clustering, a cBot Cluster Generation System with the TruSeq PE Cluster Kit v3-cBot-HS (Illumia) was employed by following the manufacturer’s instructions. After cluster generation, library preparations were sequenced on an Illumina Novaseq platform, generating 150 bp paired-end reads. To quantify gene expression, Feature Counts vl.5.0-p3 was employed to count the number of reads mapped to each gene. Subsequently, the FPKM (Fragments Per Kilobase of transcript per Million mapped reads) value for each gene was calculated, considering both the gene length and the count of reads mapped to the gene. FPKM is a commonly used method for estimating gene expression levels as it simultaneously accounts for sequencing depth and gene length. The DESeq2package (v1.16.1) was utilized to determine differential expression. DESeq2 employs a statistical model based on the negative binomial distribution to analyze digital gene expression. Genes with p value < 0.05 based on DESeq2, were considered as differentially expressed genes.

The Gene Ontology and Kyoto Encyclopedia of Genes and Genomes Enrichment Analysis

We conducted pathway analysis using the identified DEGs to explore the functional changes induced by LIPUS. We utilized the Gene Ontology (GO, http://www.geneontology.org/) database, which classifies gene functions into three categories: biological process (BP), cellular component (CC), and molecular function (MF). Enriched GO terms with a p-adj < 0.05 were deemed to be significantly enriched. The top 10 enriched terms for the PFC and hippocampal region of all categories were shown in result. Different colors represent different GO functional categories, and the number at the end of each bar represents the number of differential genes enriched in the GO category.

KEGG enrichment analysis was performed to screen for a significantly enriched KEGG pathway based on p-value < 0.05 and count > 1 criteria. The level of enrichment in KEGG pathways was assessed by considering the total number of genes, the Q value, and the count of genes that were enriched within the pathway. The Rich factor is defined as the proportion of differentially expressed genes within a pathway entry relative to the overall number of annotated genes in that pathway entry. A higher Rich factor indicates a greater level of enrichment. The Q value represents the p-value following correction for multiple hypothesis testing and ranges between 0 and 1. Q value closer to zero denotes a more significant level of enrichment.

Real-Time Reverse Transcription Polymerase Chain Reaction (RT-PCR)

RNA was isolated from tissues using TRIzol reagent (Takara, Dalian, China), followed by DNase I treatment. Using reverse-transcription enzymes (Takara 6210A, Dalian, China), 1 μg of total RNA from each sample was subsequently reverse-transcribed into cDNA. The primer sequences used in the Quantitative real-time PCR (qRT-PCR) analysis are shown in Table 1. The qRT-PCR reactions were conducted using SYBR Premix Ex Taq (Takara) and an ABI StepOnePlus Real-Time PCR instrument. Briefly, each 20 μL qRT-PCR reaction contained 0.2 μg of cDNA, 10 μL of 2 × SYBR reaction mix, and 0.4 μM of the appropriate primers. The thermocycler settings were as follows: initial denaturation at 95 °C for 30 s, followed by 40 cycles of denaturation at 95 °C for 10 s and annealing/extension at 60 °C for 30 s. In the process of DNA amplification, qRT-PCR uses fluorescent dyes to detect the total amount of product after each PCR cycle. qRT-PCR possesses higher specificity, providing precise and quantitative data that can reflect the biology of the tested experimental parameters, therefore, it is commonly used to validate important results in transcriptomics.

Table 1.

Sequences of the utilized primers

Gene	Sequence
β-actin F	GGAGATTACTGCCCTGGCTCCTAGC
β-actinR	GGCCGGACTCATCGTACTCCTGCTT
hemoglobin beta F	GCACCTGACTGATGCTGA
hemoglobin beta R	TCCAAGGGTAGACAACCAG
ATP synthase peripheral stalk-membrane subunit b F	TCAACAGGGAGAAGGCACA
ATP synthase peripheral stalk-membrane subunit b R	ATGGTAGTCCAGGCGATT
beta-globin F	GATGCTGAGAAGGCTACTGT
beta-globin R	TGATAGCAGAGGCAGAGGA
hemoglobin alpha F	GTGGTGAATATGGCGAGGA
hemoglobin alpha R	TCTTGCCGTGAGCCTTGA
growth hormone 1 F	AATGCCCAGGCTGCGTTCT
growth hormone 1 R	GATGAGCAGCAGCGAGAA
Pgk1 phosphoglycerate kinase 1 F	TGGAACGGTCCTGTTGGG
Pgk1 phosphoglycerate kinase 1 R	CAGCAAGTGGCGGTGTCT
Gck glucokinase F	GATGAAAGCTCAGCGAACC
Gck glucokinase R	CCACTTGTGACACGAAACG
claudin 2 F	AGCGAACAGGTTCCGAAGA
claudin 2 R	GCTGGCACCAACATAAGA
Occludin F	TGCTGGTTGCTGGAGAAG
Occludin R	GAGACAGGAAACGGATGG

Statistical Analysis

Three independent biological replicates were evaluated for each sample. The relative gene expression was determined using the 2-ΔΔCt approach, with β-actin serving as the internal control. One-way analysis of variance (ANOVA) and GraphPad prism 7 Software were conducted to compare groups in terms of animal behavior and qRT-PCR results. Statistical significance is indicated by asterisks (*p < 0.05, **p < 0.01). The figures show the means and the error bars indicate SEMs.

Results

LIPUS Alleviated Depression-Like Behaviors in CUS Rats

Consistent with previous works, LIPUS significantly reduced the state of anhedonia, evidenced by significant increase in the percentage of sucrose water intake of CUS rats with LIPUS. Such improvement was also observed in FST and LIPUS rats experienced less immobility time than CUS, but there was no difference in the open field test. The results have been published in our previous article [11]. Furthermore, we employed two behavioral experiments to verify the effect of LIPUS on depression. In NSFT, there was no significant difference in the latency time of animals’ head exploration of food in the area, but the latency time of animals’ beginning to bite food in the CUS group was significantly longer than that in the control and LIPUS groups (Fig. 1B, C). The LDT was employed to assess rodent anxiety by recording the number of shuttle times. After the first day of box adaptation and shuttle leading, the latency time of the CUS group was significantly longer than that of the LIPUS and the control groups on the second day. The number of shuttle times was highest in the LIPUS group, followed by the control group and the CUS group. Collectively, these results suggested that LIPUS successfully induced antidepressant-like effects in the current study.

LIPUS and CUS Altered Transcriptome Profile of the Hippocampus and Prefrontal Cortex

Previously published studies have indicated that LIPUS can effectively mitigate anhedonia and despair in CUS rats, evidenced by increased sucrose consumption in the sucrose preference test and reduced immobility time in FST. In this study, we measured the mRNA expression profiles in the PFCs and hippocampus, two brain regions closely associated with the pathogenesis of depression. The number of differentially expressed genes (DEGs) in each group is shown in Table 2.

Table 2.

Statistics of the DEGs

Comparisons	Hippocampus		Prefrontal cortex
	Up	Down	Up	Down
Ctrl vs. CUS	1805	1659	923	858
Ctrl vs. LIPUS	1583	1304	1466	1482
CUS vs. LIPUS	897	700	1029	1081

Ctrl, control group; CUS, chronic unpredictable stress group; LIPUS, low-intensity pulsed ultrasound group

Compared to the control group, the CUS group exhibited 3464 DEGs in the hippocampus, including 1805 upregulated and 1659 downregulated genes (Fig. 2A). Following ultrasound stimulation, we observed significant differential expression of 1597 genes in the LIPUS group compared to CUS rats, comprising 897 upregulated and 700 downregulated genes (Fig. 2B). In the PFC, we identified 1781 DEGs when comparing the CUS with the control group (Fig. 2C). Similarly, 2110 DEGs were identified when comparing the LIPUS with the CUS groups (Fig. 2D). Among the DEGs, 923 genes were upregulated and 858 were downregulated in the CUS group, while 1029 genes were upregulated and 1081 genes were downregulated in the LIPUS group. Venn diagrams show the overlaps of the regulated genes (Fig. 2E, F). Notably, in the hippocampus, ultrasound stimulation reversed the expression of 592 genes, with 359 upregulated and 233 downregulated genes in the LIPUS group. Similarly, in the PFC, ultrasound stimulation reversed the expression of 254 genes in the CUS group, of which 118 were upregulated and 136 were downregulated (Table 3).

Fig. 2.

Volcano plots of the differential mRNAs in the hippocampus and cortex of control, CUS and LIPUS groups. The x-axis represents the log2fold change. The y-axis represents the statistical significance (p-adj). Green and orange dots represent downregulated genes and upregulated genes respectively, and the blue dots represent genes without change. (A) and (C) show the deferential expression between the CUS group and the control group in the hippocampus and prefrontal cortex, and (B) and (D) show the changed gene expression between the LIPUS group and the CUS group in the hippocampus and prefrontal cortex. Venn diagram of the DEGs in the hippocampus (E) and prefrontal cortex (F). Yellow circle: DEGs between control and CUS; purple circle: DEGs between CUS and LIPUS. Blue circle: DEGs between control and LIPUS

Table 3.

The gene changes induced by LIPUS, compared to CUS

Position	Down in CUS but up in LIPUS	Up in CUS but down in LIPUS
Hippocampus	359	233
Prefrontal cortex	136	118

CUS, chronic unpredictable stress group; LIPUS, low-intensity pulsed ultrasound group

Functional Analysis of DEGs in LIPUS and CUS Rats

Downregulated genes in the hippocampus of depressed rats were predominantly involved in catalytic activity, oxygen carrier activity, and cofactor binding in the MF category (Fig. 3A). Conversely, the upregulated genes in the MF category were associated with GTPase binding and protein binding. The differentially expressed mRNAs in the BP category were mainly engaged in catabolic processes and regulation of neuron projection development. In terms of the CC category, the differentially expressed mRNAs were associated with nuclear speck and envelope (Fig. 3B). Following ultrasound stimulation, LIPUS rats exhibited 897 upregulated genes compared to the CUS group. In the BP category, the enriched genes were significantly associated with oxygen and gas transport. In the CC category, most of the genes were anchored components of the synaptic membrane and plasma membrane, while some were involved in oxygen carrier activity and oxygen binding in the MF category (Fig. 3D). Additionally, we analyzed 700 downregulated genes in the LIPUS group. Bases on GO enrichment analysis, these genes were associated with various biological processes, including epithelial cell migration, ameboid-type cell migration, tissue migration, and substrate adhesion-dependent cell spreading (Fig. 3C).

Fig. 3.

Enrichment analysis of the Gene Ontology (GO) pathway terms for DEGs in the hippocampus (A–D) and prefrontal cortex (E–H). The top ten enriched terms from the CUS and LIPUS groups in terms of biological process (BP), cellular component (CC), and molecular function (MF) were identified. Orange bar: biological process; green bar: cellular component; blue bar: molecular function

In the PFC, we analyzed the top ten enriched terms of downregulated and upregulated genes in the CUS and LIPUS groups (Fig. 3E–H). Compared to the control group, the downregulated genes of depressed rats were primarily associated with organic anion transmembrane transporter activity and anion transmembrane transporter activity in the MF category. In terms of biological processes, these genes were involved in sterol metabolism, sterol biosynthesis, secondary alcohol biosynthesis, and antigen processing and presentation in the BP category (Fig. 3E). Conversely, the upregulated genes in the PFC were associated with protein folding, the establishment of protein localization to organelle, and response to the topologically incorrect protein in the BP category. In the MF category, these genes were involved in unfolded protein binding, and heat shock protein binding. In the CC category, they were associated with ribonucleo protein granules (Fig. 3F).

After ultrasound stimulation, we identified 1029 upregulated genes and 1081 downregulated genes in the PFC. The downregulated genes were associated with functions such as dendritic spine membrane, C₂H₂ zinc finger domain binding, metallopeptidase activity, cell adhesion molecular binding, and heart development (Fig. 3G). On the other hand, the upregulated genes were involved in defense response to other organisms, response to viruses, innate immune response, production of molecular mediators of the immune response, immunoglobulin production, double-stranded RNA binding, and others (Fig. 3H).

We employed Kyoto Encyclopedia Genes and Genomes (KEGG) database to identify the enriched pathways with p-adj < 0.05. In the hippocampus, the CUS group exhibited several inhibited pathways compared to the control group, including amino sugar and nucleotide sugar metabolism, spliceosome, pentose phosphate pathway, and steroid biosynthesis (Fig. 4A). Conversely, several pathways were activated in the CUS group, including protein processing in endoplasmic reticulum, longevity regulating pathways (multiple species), the Forkhead box, class O signaling pathway, autophagy, longevity regulating pathway, and Rap1 signaling pathway (Fig. 4B). By comparing the LIPUS group and the CUS groups, we found that ultrasound stimulation downregulated several pathways in the hippocampus, including proteoglycans in cancer, transforming growth factor-β signaling pathway, mitogen-activated protein kinase (MAPK) signaling pathway, phosphoinositide 3-kinase-protein kinase B (PI3K-Akt) signaling pathway, focal adhesion, and amino sugar and nucleotide sugar metabolism (Fig. 4C). On the other hand, ultrasound stimulation activated several pathways such as African trypanosomiasis, malaria, terpenoid backbone biosynthesis, leishmaniasis, circadian entrainment, and steroid biosynthesis (Fig. 4D).

Fig. 4.

KEGG pathway analysis indicated the top 20 pathways with differential activity between CUS and Control and between LIPUS and CUS in the hippocampus. A Pathways with decreased activity in the CUS group compared to the control. B Pathways with increased activity in the CUS group compared to the control. C Pathways with decreased activity in the LIPUS group compared to the CUS group. D Pathways with increased activity in the LIPUS group compared to the CUS group

In the PFC, the KEGG pathway analysis revealed that the CUS group exhibited inhibition of several pathways such as steroid biosynthesis, Epstein-Barr virus infection, type1 diabetes mellitus, retinoic acid-inducible gene-1 (RIG-1)-like receptor signaling pathway, ATP binding cassette transporters, and Parkinson’s disease (Fig. 5A). The upregulated genes in the CUS group were enriched in pathways including protein processing in the endoplasmic reticulum, antigen processing and presentation, mRNA surveillance pathway, RNA degradation, and circadian rhythm (Fig. 5B).

Fig. 5.

KEGG pathway analysis indicating the top 20 differential activity pathways between CUS and Control and between LIPUS and CUS in the prefrontal cortex. A Pathways with decreased activity in CUS compared to Control. B Pathways with increased activity in CUS compared to Control. C Pathways with decreased activity in LIPUS compared to CUS. D Pathways with increased activity in LIPUS compared to CUS

Ultrasound stimulation downregulated pathways such as the MAPK signaling pathway, PI3K-Akt signaling pathway, and human papillomavirus infection in the PFC (Fig. 5C). Conversely, ultrasound stimulation activated pathways including the nucleotide oligomerization domain (NOD)-like receptor signaling pathway, Herpes simplex virus 1 infection (HSV-1), Epstein-Barr virus infection, Kaposi sarcoma-associated herpesvirus infection, hepatitis C, and influenza A (Fig. 5D). Zhang et al. used individual genotypic and phenotypic data utilizing the UK Biobank to assess the genetic effect of HSV-1 on the risk of depression, which may be regulated by several genes associated with neurodevelopment or immune function [12]. In addition, individuals with depression have altered levels and patterns of antibodies against EBV antigens. This atypical response can illuminate the immunopathogenesis of depression [13].

Verification of DEGs under LIPUS Treatment in the CUS Model

We conducted qRT-PCR analysis on a selected set of ten RNAs to validate the results of RNA-sequencing. Firstly, we ranked the genes with the most significant differences among all differentially expressed genes based on the following criteria:1. log₂fold change > 1 or log₂fold change < −1;2. p-adj < 0.1. Subsequently, within this subset, we identified six genes that LIPUS could reverse CUS changes, namely hemoglobin subunit β (Hbb), hemoglobin subunit α (Hba), β-globin, ATP synthase peripheral stalk-membrane subunit b (ATP5BP), hypothetical protein LOC688459, and growth hormone 1 (GH-1). Additionally, we selected four more genes: two of which are tight junction proteins, claudin2 (Cldn2), occludin (Ocln), known to be influenced by ultrasound on the blood–brain barrier [14], and the other two are key genes in glucose metabolism, glucokinase (GCK), phosphoglycerate kinase 1 (PGK1), as we confirmed significant changes in the glucose metabolism pathway under ultrasound in the previously mentioned enrichment pathway.

qRT-PCR exhibited high expression levels of Cldn2, Ocln, GH-1, GCK, and hypothetical protein LOC688459 in the hippocampus of CUS rats, all of which were reversed after LIPUS treatment. Furthermore, LIPUS reversed the reduced expression levels of Hbb, ATP5BP, and Hba in CUS rats. No significant changes were observed in the expression of PGK1 following CUS, but LIPUS downregulated its expression of PGK1 (Fig. 6A).

Fig. 6.

Quantitative RT-qPCR analysis of selected differential genes in the hippocampus A and prefrontal cortex B from the control, CUS, and LIPUS groups. Cldn2, claudin2; Ocln, occludin; Hbb, hemoglobin subunit β; ATP5BP, ATP synthase peripheral stalk-membrane subunit b; Hba, hemoglobinα; LOC688459, hypothetical protein; GH-1, growth hormone 1; PGK1, phosphoglycerate kinase 1; GCK, glucokinase. * means CUS vs. Control, *: p < 0.05, **: p < 0.01, ***: p < 0.001. # means LIPUS vs. CUS, #: p < 0.05, ##: p < 0.01, ###: p < 0.001. ns, not significant

In the PFC, qRT-PCR showed that LIPUS markedly reversed the expression of Hbb, β-globin, Hba, LOC688459, GH-1, PKG1, and GCK. Although ATP5BP expression in the CUS group did not show a significant difference compared to the control group, LIPUS significantly improved its expression in the PFC. Compared to the control group, Cldn2 and Ocln were downregulated in the CUS group. The expression of these genes was further downregulated by LIPUS in the PFC (Fig. 6B). Notably, LIPUS reversed the expression of Hbb, GH1, and GCK in both the hippocampus and PFC.

Discussion and Limitation

We previously demonstrated that LIPUS applied to the ventral medial PFC can effectively alleviate depressive-like behaviors in the model of CUS. In this study, we sought to unravel the underlying mechanisms by conducting a transcriptome analysis to investigate the changes induced by LIPUS at the molecular level. Interestingly, we observed significant differential expression of genes in response to ultrasound stimulation. Out of 1597 DEGs in the hippocampus, the expression of 592 genes were reversed after LIPUS. Similarly, in the PFC, 254 genes out of 2110 DEGs in the CUS group were reversed by LIPUS. The functional analysis provided insights into the similarities and differences between the hippocampus and PFC in terms of the antidepressant effects of LIPUS. One common feature was a stress-induced reversal of downregulated steroid biosynthesis in both brain regions. Enrichment analysis of DEGs revealed that stress significantly reduced oxygen carrier activity in the hippocampus, which was reversed by LIPUS. Additionally, LIPUS downregulated the adhesion function of endothelial cells in the hippocampus. In contrast, in the PFC, LIPUS significantly enhanced the immune response while reducing the adhesion function of endothelial cells.

The hippocampus and PFC are two limbic forebrain regions with distinct roles in depression. The hippocampus, which is highly susceptible to stress, serves as a central regulatory center for the HPA axis [15], while the PFC is responsible for integrating cognition and emotion [16]. Our findings suggest that there are differences in the gene-level response to LIPUS between the hippocampus and PFC, indicating potential regional variations in the mechanisms underlying LIPUS-mediated alleviation of depression.

First, we found that LIPUS significantly enhanced the oxygen carrier activity in the hippocampus, which could contribute to its antidepressant effect. This finding aligns with the upregulation of Hba and Hbb expression in qRT-PCR analysis. Hba and Hbb, along with heme, form hemoglobin, which is crucial for oxygen transport in mammalian blood. Reduced peripheral hemoglobin levels have been associated with anemia. Several cross-sectional and longitudinal studies have identified a correlation between anemia and depression [17, 18]. Furthermore, low hemoglobin levels have been correlated with fatigue and decreased brain energy [19], both of which are common symptoms in depressed individuals. However, the exact mechanism underlying this relationship is not fully understood. It is hypothesized that anemia affects the function of the brain, an organ that consumes a significant amount of oxygen and energy.

We observed a decreased expression of Hba and Hbb in the hippocampus of CUS rats, a trend that is consistent with abnormal peripheral hemoglobin levels in individuals with depression. Notably, LIPUS significantly increased the expression of Hba and Hbb, nearly quadrupling their levels. These results not only corroborate the therapeutic effect of LIPUS but also delineate a potential mechanism underlying increased hemoglobin production and subsequent improvements in cerebral oxygenation. Our results are in some agreement with the study of Kim et al. who monitored cerebral hemodynamic changes during transcranial ultrasound stimulation (tUS) by optical intrinsic signal imaging. It was found that tUS of the cortex was able to induce an increase in oxygenated hemoglobin [20].

Interestingly, in the PFC, we observed an inverse association, where LIPUS downregulated the expression of Hba and Hbb. This discrepancy may be attributed to the direct stimulation of the PFC by LIPUS, triggering compensatory mechanisms that decreased hemoglobin subunits. These findings underscore the selective impact of stress on the hippocampus and PFC, resulting in differences in cerebral oxygenation between these two brain regions.

Second, we observed that LIPUS downregulated the adhesion function of endothelial cells in both the hippocampus and PFC, while upregulating immune function specifically in the PFC. This suggests the differential effect of the immune response induced by LIPUS between the two brain subregions. Cldn2 and Ocln, are crucial components of tight junctions between endothelial cells, and the regulation of cell–cell contact and mucosal barrier function. The expression of these two proteins typically reflects the adhesion function of endothelial cells [21]. Although qRT-PCR showed notable downregulation of Cldn2 and Ocln in both the hippocampus and PFC in the LIPUS group, their expression patterns exhibited opposite patterns in response to stress in these brain regions, further highlighting the brain area-specific changes in the response to stress. Abnormal immune function and inflammatory reactions play a prominent role in the progression of depression [22]. However, our enrichment analysis revealed that LIPUS-induced reversal of immune overactivity observed in the CUS model, specifically occurred in the PFC but not in the hippocampus. This suggests potential differences in the sensitivity of different brain regions to inflammatory responses. Although some studies have demonstrated consistent stress-induced activation of glial cells and release of inflammatory factors in both brain regions, others have shown milder changes in gene expression patterns in the hippocampus compared to the PFC. For example, Posillico et al. found no significant effects or interactions of the reproductive state and/or stress on the expression of CD11b, IL-1β, IL-6, or BDNF in the hippocampus after pregnancy or FST stress, compared to the mPFC [23]. The differential effects of candidate drugs in alleviating depressive-like behavior by modulating cytokine release and neuroinflammation in specific brain regions may be attributed to the heterogeneity of glial cells in these specific brain subregions [24].

Third, we observed that DEGs induced by CUS were significantly enriched in sugar metabolism and protein processing, which are known to play a critical role in promoting stress-induced depression [25]. However, the enrichment analysis of LIPUS-induced DEGs did not indicate their significance in these pathways. Accordingly, we selected several relevant genes for PCR verification to investigate whether LIPUS can ameliorate depressive symptoms by modulating the disrupted function of sugar metabolism and protein processing. Our results demonstrated consistent changes in the expression levels of GH-1 and GCK in both brain regions. These genes were significantly upregulated in the CUS group compared to the control group, but their expression was normalized after treatment with LIPUS. GH-1, a peptide hormone that promotes protein synthesis, plays a significant role in sleep, cognition, mood, and neuroprotective effects by binding to GH receptor-1. It acts as a neurotrophic factor involved in synaptic function, neural survival, cell proliferation, and anti-oxidative stress [26]. GCK, on the other hand, catalyzes the conversion of glucose to glucose-6-phosphate, serving as the starting point of glycolysis. The consistent alteration of GCK in both the hippocampus and PFC suggests that glucose sensing and neural control of metabolism do not significantly differ between these brain regions, emphasizing its crucial role in regulating glucose homeostasis in the brain. In contrast, the level of PKG-1, a key kinase in glycolysis, was only reversed by LIPUS in the PFC and not in the hippocampus. Abnormal levels of PKG1 are associated with cancer, promoting cell metabolism and sugar metabolism [27]. These findings suggest that LIPUS may modulate the abnormal conditions in glucose metabolism and protein synthesis to elicit its antidepressant effect. Furthermore, the differential response of PKG1 suggests the presence of region-specific regulatory mechanisms in the brain.

The primary pharmacological mechanism of currently effective antidepressants involves the modulation of neurotransmitter changes. However, by investigating the antidepressant-like effects of LIPUS on mice, we did not observe the enrichment of pathways related to neurotransmitters in the differentially expressed genes identified in the transcriptomic analysis. Despite our previous study, which showed that LIPUS can modulate neural activity and the expression of molecules such as BDNF, these findings were not reflected in our transcriptomic results. This suggests that a more comprehensive study is needed to investigate the biochemical and immunological aspects of depression indicators. The lack of emphasis on these pathways may be a limitation in fully understanding the molecular underpinnings of LIPUS’s potential antidepressant effects. Future studies should address these gaps by incorporating a broader range of biological assays to explore the complex interplay between LIPUS and the pathogenesis of depression.

Conclusion and Future Research Direction

Our published findings have shown that LIPUS can significantly improve depression-related symptoms in rats subjected to stressors. Using GO and KEGG analysis and validation through qRT-PCR, we speculated that the antidepressant effect of LIPUS may rely on oxygen carrier activity, glycolipid metabolism, and immune function, which collectively promote neural activity and exert a neuroprotective effect.

The transcriptome data obtained from LIPUS-treated rats offer invaluable mechanistic insight for expanding the application of ultrasound, a category in neuromodulation therapy, to the clinical treatment of depression. On the one hand, we propose that promoting hemoglobin production and subsequent improvement in cerebral oxygenation may improve depression. On the other hand, we found that the modulatory effects of LIPUS are brain-region specific, suggesting that the optimal stimulation location of LIPUS needs further clarification. The noninvasive neuromodulation effects of LIPUS for neuropsychiatric disorders are still in the preclinical stage, and its clinical safety and efficacy in the treatment of depression need to be explored.

Author Contributions

S.R and Z.G contributed to the investigation, validation, writing- original draft preparation, and visualization. J.Z processed the methodology and resources. Y.H contributed to review & editing and investigation. Z.S and J.Y instructed all experiments, supervised this study and critically revised this manuscript.

Funding

This study was supported by the National Natural Science Foundation of China (82171525, 82171526), Beijing Municipal Administration of Hospitals’ Youth Programme (QML20181901), and Beijing Young Top-Notch Talent Support Project (2018000021223ZK36).

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Ethics Approval

The study was approved in line with the animal ethics requirements of the Animal Use and Care Committee at Capital Medical University (AEEI-2018–003).

Competing Interests

The authors declare no competing interests.