-

Neonatal inflammation leads to the development of depression-like behavior only under aversive conditions in adult males.

Depression is a general mental disorder that can be either a syndrome of another disease or an independent illness: major depressive disorder (MDD). According to the WHO, >7% of adults have a diagnosis of depression, and people at ages 12–25 have the highest prevalence rates according to data from the National Institute of Mental Health (NIMH; Center for Behavioral Health Statistics, 2021)¹. The population of patients with depressive behavior is heterogeneous with respect to symptoms, treatment responses, and biological correlates. The pathogenesis of depression includes genetic predisposition, disturbances in the synthesis and activities of monoamines in the central nervous system (CNS), aberrations in the functioning of the hypothalamic–pituitary–adrenal (HPA) axis, alterations in brain regions/functions, and long-term low-grade neuroinflammation (Beijers et al., 2019).

More than 30 years ago, it was demonstrated that inflammation plays a substantial role in the pathogenesis of depression (Maes et al., 1990), especially its treatment-resistant subtype (Dowlati et al., 2010). Several “inflammatory” models of depression have been devised, such as immune activation during pregnancy (Patterson, 2011; Bilbo et al., 2018) or during the early postnatal period (Lucchina et al., 2010; Liang et al., 2019; Cao et al., 2021) or administration of proinflammatory cytokines (Udina et al., 2012); these models have revealed an effect of immune-system activation on the pathogenesis of neuropsychiatric disorders. Moreover, even noninfectious stress activates the immune system and triggers “sterile” inflammation (Iwata et al., 2013) associated with changes in neurotransmitter and neuropeptide systems. Various meta-analyses have proven that in approximately a third of patients with depression, there are biological signs of peripheral inflammation that disappear during an antidepressant treatment, e.g., psychotherapy. Among inflammatory markers, elevated levels of C-reactive protein, IL-6, and TNF are the most reliable markers of depression (Haapakoski et al., 2015; Enache et al., 2019). The latest trial showed that people with recurrent, especially treatment-resistant depression, also feature an elevated level of IL-17, which can become an additional diagnostic marker of treatment-resistant depression (Gałecka et al., 2021; Nothdurfter et al., 2021). Clinical trials suggest that when the inflammation level is high initially (C-reactive protein >5 mg/dL), combined use of cytokine antagonists (e.g., TNF antagonist infliximab) and antidepressant therapy can improve the course of depression (Raison et al., 2013).

Newborns are very susceptible to various types of infections owing to the continued maturation/formation of various organs and systems of organs, including the immune system (Yu et al., 2018). A specific feature of the immune system of newborns is the immature status of adaptive immune responses. Therefore, innate immune responses are crucial for survival in the early neonatal period (Yu et al., 2018). It is known that some biomolecules (cytokines, adhesion molecules, and the complement system) participating in innate immunity are expressed in the healthy brain and play an important part in neurogenesis, neuronal migration and differentiation, synapse formation, and neuroplasticity (Irwin and Cole, 2011). Accordingly, either deficiency or overactivation of the immune system early in life can potentially have long-term detrimental effects on brain development and behavior. One of the possible reasons is persisting low-grade inflammation in the body and brain, which in stressful situations in adulthood can have the same impact as direct activation of the immune system. In longitudinal studies, perinatal infection is reported to raise the risk of mood disorders, including anxiety and MDD (Chu et al., 2019; Toenders et al., 2021). In a large-scale study called the Avon Longitudinal Study of Parents and Children (ALSPAC), researchers found that early-childhood infections correlate with the development of depressive symptoms in mid-adolescence (Chaplin et al., 2020). Children with elevated levels of IL-6 and C-reactive protein at 9 years of age are at a higher risk of depression onset by 18 years of age (Khandaker et al., 2018). Epstein–Barr virus exposure at age 4 years is associated with subsequent risk of definite psychotic experiences in adolescence (Khandaker et al., 2014).

Investigation into rodent models confirms that inflammatory processes during the embryonic and neonatal period can result in the development of neuropsychiatric disorders later in life (Meyer et al., 2009; Liang et al., 2019; Cao et al., 2021; Khantakova et al., 2022a). The mechanism of neuronal anomalies during maternal immune activation involves impairments of active development of brain structures (Gałecka et al., 2021). On the other hand, molecular mechanisms of the neurobehavioral disorders that emerge after neonatal immune activation (NIA) are not yet fully understood. In 2012, in their review, Bilbo and Schwarz comprehensively examined the delayed impact of perinatal infection on immune responses and cognition later in life (Bilbo and Schwarz, 2012). Those authors distinguished the direct and indirect effects of immune activation on neuronal functions: (1) the direct influence of immune activation is the introduction of irreversible alterations or disturbances into certain neural processes, e.g., neurotransmission and synaptic plasticity, which is key to learning and memory; and (2) by the indirect effect, those authors mean a reprogramming of the immune function of the adult human body, i.e., a change in how a response to a subsequent immune challenge takes place in adulthood (Bilbo and Schwarz, 2012). Indeed, there is growing evidence of relations among low-grade inflammation, levels of proinflammatory cytokines, and changes in neurotransmitter and neuroendocrine functioning in psychiatric disorders, during depressive and anxious behavior, and in MDD (Chu et al., 2019; Toenders et al., 2021).

One of the simple experimental approaches to the research on the influence of the immune system on the developing brain is to induce antibacterial immune responses without direct administration of bacterial pathogens, i.e., via injection of bacterial endotoxin: lipopolysaccharide (LPS). The use of the mimetic rather than a live bacterium has a number of advantages: biological safety, dose-dependent intensity of the immune response (Furube et al., 2018), and the ability to quantify cytokines (Yan and Kentner, 2017; Custódio et al., 2018; Berkiks et al., 2019a) and the expression of immune-system–related genes (Cao et al., 2021; Kvichansky et al., 2021) in the blood and various tissues, including the brain. Additionally, the effect of mimetic administration is time-limited: 24–48 h depending on the dose (Furube et al., 2018). These advantages allow us to set the exact time of immune activation corresponding to specific periods of neuroimmune-axis development; this approach is especially relevant for studies on the influence of an immune challenge on brain dysfunction (Liang et al., 2019; Luo et al., 2021; Zhao et al., 2022). It is worth noting that mimetics cannot simulate the full range of immune responses caused by a bacterial infection; they mimic only cytokine-related responses of the acute phase of inflammation (Schwarz and Bilbo, 2011).

Overall, it is now known that immune activation at an early age can lead to impaired brain development and may influence behavior later in life. Nevertheless, the limitations of human studies and the heterogeneity of animal models make it difficult to pinpoint the impact of neonatal inflammation on the risk of depression at an older age. To address this issue, here we analyze the results of animal studies involving a model of LPS-induced neonatal inflammation in rodents (mice and rats). Besides, we discuss long-term molecular changes in brain structures and their relations with depressive-like behavior in adolescents and adults.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed during the preparation of this review. Two authors (JK and EA) searched Scopus databases independently to select the articles that would best reflect the assumptions of the study in the “Introduction” Section. “Neonatal immune activation” was combined with the term “depression, ” “forced swim, ” “tail suspension, ” “anhedonia, ” and “rodent” to find relevant articles (title/abstract/key). The present authors included only original articles in the English language that were related directly to LPS in NIA. The databases were searched in July–August 2022. The authors retrieved 239 records from Scopus (Supplementary Material) and excluded articles having the year of publication before 1990.

After an initial review, 30 articles were excluded because they were review articles, editorials, commentaries, letters to the editor, or not full-length article. Next, 154 articles, involving maternal/prenatal/in utero immune activation, administration of a mimetic after postnatal day 14, early-life social stress (separation), ischemia, alcohol or other mimetics, or cell culture research were excluded from the study. Eight studies on other types of behavior (e.g., pups’ behavior, anxiety, memory, learning, and locomotor activity) as well as studies on a long-term molecular change without an analysis of behavior were also excluded from the analysis (Supplementary Material). Thus, we selected 25 full-length animal research articles for the analysis, all related to the development of depressive-like behavior in adolescents (postnatal days 30–45) or adult rodents (after postnatal day 60) after neonatal administration of LPS and associated molecular patterns. In three studies, their authors investigated long-term alterations in both age groups.

The consequences of NIA in adolescents are studied in eight research articles, whereas in adults, in 20 studies, among which, three articles (Dinel et al., 2014; Tishkina et al., 2016; Custódio et al., 2018) address long-term alterations in both age groups of animals. In 10 (40%) studies, experiments are conducted on different strains of mice, and in 15 (60%) studies, on different strains of rats. In eight (32%) studies, investigators use only males, in 13 (52%) studies, males and females are analyzed separately, and in four (16%) studies, there are mixed groups. In 40% of our selected studies (Table 1), researchers employ the “dual-exposure-to-endotoxin” model of mimetic administration, where different doses of endotoxin (0.025–0.05 mg/kg) are administered twice: on postnatal days 3 and 5. In eight (32%) studies, an endotoxin (0.05–0.1 mg/kg) is administered once on postnatal day 14, and in one study (4%) twice on postnatal day 14–15, already at the end of the stress-hyporesponsive period. Furthermore, in six (24%) studies, 0.025–2.0 mg/kg LPS is administered at various time points during the first 2 postnatal weeks. Nonetheless, regardless of the administration scheme, the administration of LPS drives the activation of microglia and astrocytes (Zhao et al., 2022), upregulation of mRNA expression of Il-6, Il-1β, and Tnf in various brain regions, and of respective protein levels in blood plasma (Dinel et al., 2014; Cao et al., 2021; Zhao et al., 2022), and enhanced production of corticosterone (Lucchina et al., 2010) during the first 24 h. To make the data compatible among the studies, in articles not specifying the sex of the studied animals, we assumed that investigators analyzed a mixed group of animals. To calculate the percentage of animals in which certain changes were observed, we utilized data on the number of animals presented in the original articles. In the absence of data on the numbers of females and males in mixed groups, we attributed the reported number of animals to both sexes.

To quantify depressive-like behavior in animal models, researchers commonly use the tail suspension test (Steru et al., 1985), the forced swim test (Porsolt et al., 1977), and the sucrose or saccharine preference test (Willner et al., 1987). Less commonly, the open field test, the novelty-suppressed feeding test (Dulawa and Hen, 2005), and the elevated plus maze test are employed, which primarily evaluate the general locomotor activity and anxiety-like behavior rather than depressive behavior; for this reason, studies involving such tests are not analyzed in this review.

The forced swim test is the most common assay of depressive behavior. In the forced swim test, an animal is placed in an aversive environment: a container of water from which the animal cannot escape. After unsuccessful attempts to get out, the animal stops struggling (and floats with minimal movements needed only to keep the nose above the water). The length of time of the animals’ trying to find a way out under the unavoidable aversive conditions reflects the degree of their behavioral despair and correlates well with the therapeutic action of antidepressants (Kara et al., 2018). The tail suspension test is similar to the forced swim test and also assesses the degree of behavioral despair but does not involve water immersion, which is a strong stressor and can induce hypothermia in animals. In the tail suspension test, rodents are hung by the tail, and the length of time is also determined after which the animals stop looking for a way out of the aversive situation and assume an immobile posture. The sucrose or saccharine preference test is used to evaluate the ability to experience pleasure (Willner et al., 1987). One of the symptoms of depression is anhedonia, i.e., the inability to experience pleasure. A decrease in the consumption of sucrose, which rodents normally prefer, indicates the development of anhedonia.

Analysis of the literature indicates that NIA does not lead to anhedonia. Only one of four studies on sucrose preference in both adolescent and adult rodents shows a reduction in sucrose intake (Table 1; Majidi et al., 2016; Cao et al., 2021). This indicates the absence of symptoms of depressive-like behavior in a steady state. By contrast, placement of animals with a history of NIA into aversive settings (the forced swim test or tail suspension test) leads to the manifestation of pronounced symptoms of depressive-like behavior in adult animals, as reported in 14 out of 20 studies (Table 1). Of note, males present with depressive behavior more frequently (55%–60% of animals analyzed) than females do (23%–32% of animals analyzed), implying sex-specific effects of NIA. In one work, on the contrary, a decrease in depressive behavior in adult males is demonstrated, possibly due to low sensitivity of F1 C57Bl/6 × DBA-2 hybrids to the forced swim test (Luo et al., 2021). Unexpectedly, an increase in immobility in the forced swim test or tail suspension test was observed only in three out of seven studies on adolescent rodents (postnatal days 30–48), and there are no pronounced differences between females (21%–26% of the studied animals according to data from the forced swim test) and males (27%–34% of the studied animals according to data from the forced swim test). Thus, neonatal inflammation in most cases leads to depressive-like behavior in adult males, but not in females or adolescents of both sexes, and only under aversive testing conditions.

Of note, under stable environmental conditions in both humans and rodents, it is difficult to analyze delayed behavioral symptoms after neonatal inflammation. For this reason, investigators often use various aversive conditions to assess depressive behavior and associated molecular aberrations. As aversive conditions, researchers have utilized the forced swim test or tail suspension test (Doosti et al., 2013; Berkiks et al., 2018a; Custódio et al., 2018; Kvichansky et al., 2021) or the restraint test (Majidi et al., 2016), which may reflect an effect of acute stress. Or, in the case of acute inflammatory stress, the administration of various doses of LPS to adult animals has been used (Kentner et al., 2010; Lucchina et al., 2010; Bernardi et al., 2014; Yan and Kentner, 2017). Surprisingly, in one study, after an adult LPS rechallenge, researchers documented a decrease, not an increase, in the depressive behavior of males during the forced swim test (Bernardi et al., 2014). Those authors explain this result by the induction of endotoxin tolerance in response to neonatal LPS exposure. The absence of other studies with such a design does not allow us to either confirm or refute the validity of this conclusion.

The lack of an effect of neonatal inflammation in female rodents may be due to increased variation of outcomes owing to the influence of the estrous-cycle stage on behavior. Only one of the presented studies addresses the influence of this cycle’s stage on the behavior of females (one of 17 studies for all ages, or 6%). To obtain relevant results, it is necessary to conduct more systematic studies on females, increase the sample size, and subdivide the animals into groups according to the stage of the cycle.

Thanks to advances in neuroimaging studies, brain regions have been identified that are significantly impaired in patients with MDD. In the brain, they include those areas that regulate motivation, locomotor activity, arousal, and anxiety (Miller and Raison, 2016). Various areas of the frontal cortex and limbic regions are involved in the regulation of these types of behavior, for example, the amygdala, hippocampus, and striatal regions including the caudate, putamen, and nucleus accumbens (Li B.-J. et al., 2018; Zhang et al., 2018). Neuron division in most regions of the brain is already completed by the time of birth but continues in the dentate gyrus of the hippocampus and subventricular zone. Therefore, the main negative effect of NIA is absorbed by cells that perform immunocompetent and homeostatic functions in the brain: microglia and astroglia. These cells actively divide and undergo a series of morphological alterations during the neonatal period (Bilimoria and Stevens, 2015; Abbink et al., 2019). Changes in the morphology and amounts of microglia and astroglia, just as processes of neurogenesis, are dynamic indicators of the state of the brain, and their disturbances point to pathological conditions, including neuroinflammation.

Our analysis of the literature suggests that owing to persisting molecular changes in brain tissues, early-life neonatal inflammation is a risk factor for depression-like behavior in adult males only when they face aversive/difficult life situations. Elevated oxidative stress raises glial activation and stimulates proinflammatory signaling pathways, thereby possibly prolonging/perpetuating inflammation and contributing to the abnormalities of brain functions and of neuronal signaling that are seen in depression (Bhatt et al., 2020). A cytokine imbalance and altered reactivity of the HPA axis to a second stressor—along with normal basal levels of corticosterone, TNF, and IL-1β—are consistent with the “two-hit theory” of depression development, according to which exposure to environmental stress at an early age enhances behavioral and hormonal responses to an additional stressor in adolescence and adulthood (Calcia et al., 2016; Reshetnikov et al., 2021). Furthermore, the combined effects of reactive oxygen and nitrogen species and of inflammatory cytokines reduce the availability of tetrahydrobiopterin (BH4): a key enzyme cofactor for the synthesis of all monoamines (Neurauter et al., 2008). It is also known that upregulation of IL-1β and TNF stimulates the function of serotonin reuptake transporters through IL-1R and the p38 MAPK pathway, thereby lowering the synaptic availability of serotonin and causing a manifestation of depressive behavior in laboratory animals (Zhu et al., 2010).

Back in 1993, T.B. Herbert and S. Cohen reported an association between the development of clinical depression and alterations in cellular immunity (Herbert and Cohen, 1993). In case-control studies on MDD, genes connected with a response to infection and with innate immunity have been shown to be overexpressed (Leday et al., 2018; Wittenberg et al., 2020), in agreement with data indicating activation of innate immunity and enhanced signaling of proinflammatory cytokines (Dantzer et al., 2008; Miller et al., 2009). At the same time, genes related to T-cell function and adaptive immunity are underexpressed (Leday et al., 2018), also consistently with previous findings pointing to a relative decrease in natural killer cytotoxicity and in lymphocyte proliferation under the action of mitogens in vitro (Zorrilla et al., 2001) and with recent data suggesting that natural killer and T-cell deficiency is accompanied by monocyte inflammatory activation (Grosse et al., 2016; Snijders et al., 2016).

The described molecular alterations in depressed animals with a history of NIA have pathogenetic patterns similar to those in patients with MDD and in depressive-like behavior after NIA (Table 6). This observation suggests that early-life neonatal infection causes long-term alterations of the already characterized pathogenic changes in MDD. For instance, some of the genes overexpressed in MDD are genes participating in the response to LPS (Leday et al., 2018), consistently with overexpression of TLR4 in the prefrontal cortex in depressed subjects and suicide victims (Pandey et al., 2019). This phenomenon may also be the reason for the over-reaction of cytokines during inflammation in adults after NIA.

Aside from hypersensitivity to inflammatory stimuli, anomalies in the regulation of immune responses are seen in depression. It has been shown that the depletion of regulatory T cells (CD4⁺ CD25⁺ Tregs) in vivo can lead to anxious and depressive behavior (Kim et al., 2012) via increased inflammation (Grosse et al., 2016; Snijders et al., 2016). A case-control study indicates that after effective antidepressant therapy, the number of regulatory T cells in the blood goes up and the number of T helper 17 (Th17) cells diminishes (Jahangard and Behzad, 2020). Recent research also indicates that CD4⁺ CD25⁺ regulatory T cells can inhibit inflammatory responses in depressed patients (Ellul et al., 2018). A transfer of off-spring antigen–specific regulatory T cells from the dams that had toxoplasmosis during pregnancy reverses pups’ behavioral abnormalities, thus implying the therapeutic potential of adoptive regulatory-T-cell transfer in neuropsychiatric disorders associated with immune alterations (Xu et al., 2021). This is a rare piece of direct evidence linking regulatory T cells and behavior. Furthermore, a decrease in plasma serotonin concentration correlates with the downregulation of the 5-HT1a receptor on the surface, but not in the cytoplasm, of regulatory T cells in depressed patients (Li et al., 2010). Because serotonin takes part in the activation of lymphocyte proliferation through the 5-HT1a receptor (Aune et al., 1993), the decrease in the concentration of this monoamine in the blood reduces the proliferation and the number of regulatory T cells.

The microbiome also plays a considerable role in the development and persistence of MDD (Jiang et al., 2015), mainly owing to an imbalance between Th17 cells and regulatory T cells toward the former (Westfall et al., 2021a,b). IL-17, produced by RORγt⁺ Th17 cells, contributes to the development of depression, especially its treatment-resistant subtypes (Nothdurfter et al., 2021). The literature shows that the number of Th17 cells in gut-associated lymphoid tissue increases in response to stress and correlates with depression-like behavior in mice (Westfall et al., 2021b). Besides, administered Th17 lymphocytes have been found to cause depression-like behavior in mice (Beurel et al., 2013).

Because of the similarity of molecular alterations that occur during MDD and after NIA as well as the already shown influence of the microbiota and Th17 or regulatory T cells on the development and course of depression, a question arises: what effect does neonatal inflammation have on the emerging microbiota of pups and on the establishment of a balance between Th17 cells and regulatory T cells? It is known that regulatory T cells arise in the thymus and begin to migrate to peripheral organs and tissues, primarily to the skin and intestinal walls, at early stages of embryogenesis (Scharschmidt et al., 2015, 2017) and are necessary to maintain tolerance to one’s own emerging microbiota (Mold et al., 2008; Lathrop et al., 2011; Ohnmacht et al., 2015; Westfall et al., 2021b). In newborn mice, it has been shown that the primary response to an antigen induces Th1 and Th2 immune responses, with equal production of interferon gamma (IFN-γ) and IL-4 (Mold et al., 2008). During an antigen rechallenge, rapid production of IL-4 occurs, which causes apoptosis of neonatal Th1 cells through the formation of heteroreceptor IL-4Rα/IL-13Rα1, which together with regulatory T cells takes part in the control over peripheral tolerance and restrains autoimmunity in adults (Barik et al., 2017). This observation further confirms the initial tolerogenic type of the immune system in newborns.

Overall, the phenotypic diversity of regulatory T cells allows them to access various tissues, including brain tissue, where they implement immune homeostasis (Tanoue et al., 2016; Khantakova et al., 2022b). In the CNS, regulatory T cells play a major part in the regulation of inflammation. In the rat brain, regulatory T cells constitute ~15% of CD4⁺ T lymphocytes (Xie et al., 2015). Notably, brain astrocytes may perform a regulatory function in the control of cerebral regulatory T cells by promoting the expression of Foxp3 via the IL-2–STAT5 pathway (Xie et al., 2015). Neurons also contribute to the differentiation of regulatory T cells, via B7 family proteins and the TGF-β1 pathway (Liu et al., 2006). Stressors during the critical period of immune-system formation and migration of regulatory T cells can have long-term and unpredictable consequences, by reducing their functional activity and migration potential.

Of note, the sex- and age-specific outcomes that we found were somewhat unexpected. The changes we noted in adult males (but not in females or adolescents regardless of gender) make us wonder which molecular mechanisms determine the timing of the manifestation of delayed effects of NIA and its gender specificity. Human studies show that childhood infectious diseases increase the risk of psychotic illness in adulthood regardless of gender (Dalman et al., 2008; Khandaker et al., 2012). Nonetheless, a recent population-based longitudinal birth cohort study (Chaplin et al., 2020) revealed that early-childhood infections are associated with the risk of depressive symptoms in adolescence (age 12–17 years) but not in early adulthood (age 18–19 years). The inconsistency of the results about humans may be due to the population’s genetic background or to the socioeconomic characteristics of the subjects and to the assessment methods. After interpreting the data, we can hypothesize that behavioral patterns in mice during adolescence make it difficult to detect depression-like behavior in the tail suspension test and forced swim test. It is known that adolescent rodents are less anxious than juvenile or adult rodents and are more prone to risk-taking, impulsive behavior, and novelty-seeking (Macrì et al., 2002; Stansfield and Kirstein, 2006; Arrant et al., 2013). It is possible that these features of the behavior of mice during adolescence ensure the best stress-coping behavior under aversive conditions, making differences in the manifestations of depressive-like behavior between groups insignificant. On the other hand, the scarcity of studies on adolescent and female rodents in this field may also be one of the reasons for the lack of pronounced differences. Accordingly, in the future, to assess changes in rodents during adolescence, it will be necessary to increase the number of studies and to develop specific tests for depressive-like behavior in adolescent animals.

In conclusion, our analysis uncovered delayed structural and molecular patterns linked with the development of depression-like behavior in rodents after neonatal inflammation. Due to the similarity of the identified changes with data from MDD patients, further studies should address the influence of neonatal inflammation on the differentiation and functioning of the immune system’s regulatory mechanisms as well as the search for possible ways to correct the identified disturbances.

The analysis of the impact of delayed consequences of induced neonatal inflammation in rodents has several limitations regarding the interpretation of data on neuroinflammation and depression-like behavior. The first set of limitations is related to the heterogeneity of experimental data (e.g., the lineage, mimetic administration protocol, age of analysis of behavior and molecular changes, the type of test for behavior assessment, the brain structure chosen for analysis, the extraction method, coordinates, and assessment techniques). The heterogeneous design of experiments means that we can analyze only the most pronounced changes that are common among most models but cannot examine the less pronounced effects of a particular model. Apparently, the heterogeneity of the experimental design may be the reason for the inconsistent results about the influence of NIA on both behavioral parameters and structural and molecular characteristics of neuroinflammation.

The second limitation of the animal model for evaluating the effect of NIA on depression-like behavior is related to the limited range of detectable behavior in rodents, in contrast to humans. The most common tests currently available for assessing depressive-like behavior (the forced swim test, tail suspension test, or sucrose preference test) are able to quantify only a limited range of behavioral traits associated with depression. Such symptoms as low mood and low self-esteem, worthlessness, a feeling of guilt, impaired concentration, and suicidal behavior cannot be assayed in animals but are common manifestations of depression in humans. The commonly used tests reflect specific clinical symptoms, typically associated with a lack of motivation, but do not permit assessment of deeper anomalies of mood and behavior. The forced swim test and tail suspension test involve well-pronounced aversive conditions for rodents, and according to some authors, evaluate stress-coping strategies rather than depressive-like behavior (Molendijk and de Kloet, 2015, 2019; de Kloet and Molendijk, 2016). Nonetheless, due to the simplicity and affordability as well as the complexity of alternative techniques, these tests continue to be popular in scientific research.

Finally, the third limitation is brain morphofunctional differences between rodents and humans. These differences are seen both in the neonatal period—when the stage of development of the brain in rats and mice corresponds to the stage of development of the third trimester of pregnancy in humans (Lupien et al., 2009)—and in adulthood because it is not yet possible to unambiguously match various regions of the cerebral cortex between rodents and humans. This may be the reason for the inconsistency among results of comparisons of transcriptional signatures between various mouse models of depression and postmortem data from patients with MDD (Scarpa et al., 2020; Reshetnikov et al., 2022).

Despite the limitations associated with the use of animal models of NIA, our analysis points to a close relationship between inflammation at an early age and the development of a depressive state in adults when exposed to aversive environmental conditions. This conclusion also confirms the similarity of the alterations detectable after NIA at the cellular and molecular levels with the pathological changes that proceed during depression.

JK and EA searched databases independently to select the articles. JK wrote the manuscript. EA prepared the tables. NB and VR planned and supervised the whole project. All authors contributed to the article and approved the submitted version.