Review Reports - Multisystemic Consequences of Brain-Derived Neurotrophic Factor (BDNF) Haploinsufficiency in the SD-BDNFtm1sage Rat Model

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript in well written and thoroughly reviews work done on the SD-BDNFtm1sage line, a rat genetic model of BFNF haploinsufficiency. There are several things that should be addressed:

It is important to state in the introductory paragraph that caveat of this model’s chronic BDNF deficiency is that it includes developmental changes. Authors should also point to any instances in which manuscripts assessed whether observed phenotypes are due to the developmental vs adult deficiency in BDNF levels.

Please include more information to readers about how this line was created e.g., which region of Bdnf gene was targeted in SD-BDNFtm1sage line and why is deletion of that region sufficient to decrease the BDNF expression.

Please include information on the decrease in Bdnf mRNA levels whenever it is available in mentioned manuscripts.

In a few regions where BNDF protein levels are not decreased in the SD-BDNFtm1sage rats authors state that BDNF expression is independent of the Bdnf gene e.g., “significantly upregulated BDNF and GDNF in fast-twitch muscles (tibialis anterior, medial gastrocnemius) regardless of genotype, suggesting that exercise-induced muscular neurotrophin synthesis operates independently of the Bdnf ”

Can authors please clarify this statement that (muscular) BDNF synthesis operates independently of Bdnf gene?

Authors state “ Grzelak et al. [76] also reported significant reductions in serum GDNF, NT-3, NT-4, and NGF (by 22%, 31%, 29%, and 37%, respectively), indicating broad downregulation of circulating neurotrophic support rather than an isolated BDNF deficit” Could authors discuss why are these other neurotrophic factors also reduced in serum and if this is seen in any other region in SD-BDNFtm1sage rats?

Letter size in boxes in Figure 2 is small, and it is hard to read text. Could author please include the letter size?

Author Response

Author(s): Lucyna Mrówczyńska and Włodzimierz Mrówczyński

Title: Multisystemic consequences of Brain-Derived Neurotrophic Factor (BDNF) haploinsufficiency in the SD-BDNFtm1sage Rat Model

Response to Reviewer 1

Thank you for your insightful comments, which have significantly contributed to improving our manuscript. In accordance with your suggestions, we have revised the text accordingly; all modifications are highlighted in blue. Below, we provide detailed responses to the comments raised by Reviewer 1.

Point 1

It is important to state in the introductory paragraph that caveat of this model’s chronic BDNF deficiency is that it includes developmental changes. Authors should also point to any instances in which manuscripts assessed whether observed phenotypes are due to the development vs adult deficiency in BDNF levels.

We thank Reviewer 1 for this important comment. We fully agree that this represents a crucial interpretive caveat. In the revised version of the Introduction (second‑to‑last paragraph), we now explicitly acknowledge that the phenotype observed in adult individuals reflects the combined and largely inseparable effects of BDNF‑dependent developmental disturbances and the persistent deficiency of BDNF in the mature organism.

In addition, the deletion of the Bdnf gene in these animals is constitutive, resulting in reduced BDNF levels from conception through adulthood. Consequently, the phenotype of adult heterozygous animals reflects the combined effects of BDNF-dependent developmental abnormalities and persistent BDNF deficiency in the mature brain. Given the key role of BDNF during development, chronic deficiency of this factor may induce permanent changes in the architecture of neural circuits, which in such an animal model makes it difficult to distinguish between developmental effects and those emerging in adulthood.

Point 2

Please include more information to readers about how this line was created e.g., which region of Bdnf gene was targeted in SD-BDNFtm1sage line and why is deletion of that region sufficient to decrease the BDNF expression.

We thank Reviewer 1 for this valuable suggestion. At the beginning of Section 2 (second part), we have added a dedicated paragraph describing the generation of this line and clarifying why disruption of a single region of the Bdnf gene is sufficient to reduce BDNF expression.

The SD-BDNFtm1Sage line was generated in a Sprague–Dawley rat genetic background by SAGE Labs using CompoZr zinc finger nuclease (ZFN) technology [71, 73]. A pair of ZFNs targets the Bdnf protein-coding region, inducing a double-strand break that is repaired by non-homologous end joining (NHEJ). This results in a small deletion and a premature stop codon, leading to the production of a truncated, non-functional protein instead of mature BDNF. This procedure is sufficient to disrupt all BDNF production, because although the rat Bdnf gene contains multiple alternative 5′ non-coding exons driven by separate promoters, all transcripts share a common 3′ coding exon [74]. Consequently, a mutation in this shared exon affects every transcript variant, regardless of the promoter used, preventing functional BDNF expression from the mutated allele. As a result, homozygous rodent mutants are completely depleted of BDNF and die within the first few days after birth, whereas heterozygotes retain one intact allele and survive lifelong with an approximately 50% reduction in BDNF expression [69, 70, see also Envigo technical summary of the BDNF‑KO rat model, which provides the most comprehensive overview of the homozygous and heterozygous phenotypes at: https://insights.envigo.com/hubfs/resources/data-sheets/envigo-gems-bdnf-ko-rat.pdf].

Point 3

Please include information on the decrease in Bdnf mRNA levels whenever it is available in mentioned manuscripts.

We thank Reviewer 1 for bringing this issue to our attention. Upon reviewing the literature, we found that most studies using the SD‑BDNF^tm1sage line assessed BDNF concentrations at the protein level, typically by ELISA (serum and tissues) or Western blot. To our knowledge direct measurements of Bdnf mRNA were reported by Sapio et al. [89], who performed RNA sequencing of dorsal root ganglia and dorsal spinal cord in heterozygous and wild‑type rats. Notably, these authors observed no differences in Bdnf transcript levels between genotypes, indicating that the haploinsufficiency phenotype reflects reduced functional protein rather than decreased mRNA abundance. Corresponding explanatory paragraphs have now been added to Section 2.2.

It should be noted that BDNF levels in this rat model have primarily been quantified as protein, while direct measurements of BDNF mRNA in this strain were reported by Sapio et al. [87], who performed RNA sequencing of dorsal root ganglia and the dorsal spinal cord in heterozygous and wild-type rats. However, no changes were found between heterozygous and wild‑type rats, indicating that the haploinsufficiency phenotype reflects reduced functional protein rather than decreased transcript abundance.

Point 4

In a few regions where BNDF protein levels are not decreased in the SD-BDNFtm1sage rats authors state that BDNF expression is independent of the Bdnf gene e.g., “significantly upregulated BDNF and GDNF in fast-twitch muscles (tibialis anterior, medial gastrocnemius) regardless of genotype, suggesting that exercise-induced muscular neurotrophin synthesis operates independently of the Bdnf ”

Can authors please clarify this statement that (muscular) BDNF synthesis operates independently of Bdnf gene?

We thank Reviewer 1 for pointing out this imprecise wording. The reviewer is correct that BDNF synthesis in muscle cannot be independent of the Bdnf gene, as muscular BDNF is necessarily a product of that gene. Our intended meaning was that the exercise‑induced increase in muscular BDNF concentration is independent of genotype. We have revised the relevant sentences in Section 3.5 accordingly to ensure accurate and unambiguous interpretation.

Moderate‑intensity endurance training did not restore serum BDNF levels in either genotype. However, it significantly increased BDNF and GDNF concentrations in fast‑twitch muscles (tibialis anterior and medial gastrocnemius) across both genotypes. This pattern indicates that exercise‑induced neurotrophin upregulation in skeletal muscle occurs independently of the BDNF genotype.

Contrary to earlier predictions, endurance training does not act as a universal method for restoring neurotrophin levels [92]. Instead, it activates contraction-dependent pathways that stimulate neurotrophin synthesis in skeletal muscles, and these mechanisms operate independently of the BDNF genotype remaining fully effective even in the case of heterozygous loss of a single BDNF allele.

Point 5

Authors state “ Grzelak et al. [76] also reported significant reductions in serum GDNF, NT-3, NT-4, and NGF (by 22%, 31%, 29%, and 37%, respectively), indicating broad downregulation of circulating neurotrophic support rather than an isolated BDNF deficit” Could authors discuss why are these other neurotrophic factors also reduced in serum and if this is seen in any other region in SD-BDNFtm1sage rats?

We thank Reviewer 1 for this insightful question. To address this point, we have added the relevant explanatory paragraphs to Section 2.2 of the revised manuscript.

This decrease in BDNF, GDNF, NT-3, NT-4, and NGF concentrations appears to be systemic in nature, as it was detected exclusively in serum, whereas concentrations in three hindlimb muscles (tibialis anterior, gastrocnemius, and soleus) remained unchanged [77]. The full panel of neurotrophins was not examined in any other tissue or in any other brain region in rats from the SD-BDNFtm1sage line. Since the mutation affects only BDNF, the reduction in concentrations of other neurotrophic factors is likely secondary and may result from changes in the content or release of platelets (which not only store most circulating BDNF but also transport other neurotrophins), disturbances in interactions between various neurotrophic signaling or systemic consequences of BDNF deficiency (such as HPA axis dysfunction, elevated glucocorticoid levels, obesity, etc), which may inhibit circulating neurotrophic support. Since none of these mechanisms has been investigated, the cause remains unresolved.

Point 6

Letter size in boxes in Figure 2 is small, and it is hard to read text. Could author please include the letter size?

Figure 2 has been revised, and the font size has been increased as suggested.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript addresses an important and timely question: whether chronic partial BDNF deficiency in the SD-BDNFtm1sage rat constitutes a coherent translational model of multisystem vulnerability. The topic is biologically relevant and potentially useful, given the central role of BDNF in neurodevelopment, affective regulation, sleep, nociception, reward, and metabolic homeostasis. The authors have assembled a broad literature and make a plausible case that this rat line captures more than a narrow neuropsychiatric phenotype.

The principal strength of the review is its integrative ambition. Rather than treating behavioral, metabolic, nociceptive, and neuromuscular findings as isolated observations, the manuscript attempts to organize them into a systemic model of BDNF haploinsufficiency. This is a worthwhile objective. However, the manuscript currently overstates the maturity of the evidence base. Much of the synthesis depends on small, heterogeneous preclinical studies, several of which examine different ages, sexes, interventions, tissues, and behavioral paradigms. The central claim that the model broadly reflects human multisystemic consequences of BDNF deficiency is intriguing but insufficiently qualified.

Major Comments

The manuscript needs a clearer distinction between findings directly demonstrated in SD-BDNFtm1sage rats and findings inferred from other Sprague–Dawley models or from the broader BDNF literature. Several sections incorporate studies of wild-type rats, pharmacological manipulations, or homozygous knockout controls and then fold them into the interpretive framework of the heterozygous SD-BDNFtm1sage model. This weakens conceptual precision. The authors should explicitly classify each cited study by model, genotype, intervention, sex, age, and primary outcome, and avoid implying that all mechanisms reviewed have been shown in the focal line.

The causal language is too strong. Statements suggesting that prefrontal BDNF deficiency “drives” HPA dysregulation, REM sleep disruption, hepatic pathology, and cardiovascular dysfunction are biologically plausible but not established by the cited studies. The proposed corticolimbic-systemic feedback loop is a hypothesis, not a demonstrated pathway. The manuscript would be substantially strengthened by recasting this model as an integrative framework requiring experimental validation, preferably through longitudinal designs, tissue-specific rescue, TrkB modulation, endocrine profiling, and mediation analyses.

The review gives insufficient attention to methodological heterogeneity and risk of bias. The included studies appear to vary in sample size, sex inclusion, age at testing, housing conditions, stress exposure, behavioral protocols, tissue collection timing, and assay methodology. These factors are especially consequential for BDNF, whose expression is highly sensitive to circadian timing, physical activity, stress, platelet release, diet, and sample handling. The authors should add a dedicated methodological appraisal rather than presenting the literature as uniformly convergent.

The biomarker argument requires substantial tempering. Peripheral BDNF is presented as a clinically accessible biomarker of central circuit function and broader neurobiological dysfunction. This claim is premature. Serum BDNF is heavily influenced by platelet content and peripheral sources, and cross-tissue correlations in rats do not establish clinical validity in humans. The authors should distinguish analytical validity, biological plausibility, and clinical utility, and should avoid implying diagnostic or prognostic readiness.

The translational framing is uneven. The strongest translational evidence appears to concern nociceptive thresholds, where rat findings are compared with human BDNF deletion in WAGR syndrome. By contrast, claims about depression, PTSD, schizophrenia, addiction, metabolic syndrome, and cardiovascular disease are mainly inferential. The discussion should rank translational confidence across domains rather than presenting the phenotype as equally validated across systems.

The manuscript also needs a more balanced discussion of alternative explanations. Many reported phenotypes could arise from developmental compensation, altered growth, obesity, endocrine differences, reduced brain or pituitary size, strain-specific effects, or behavioral confounding related to locomotion, motivation, and body weight. The review acknowledges developmental programming but does not consistently apply that caution across adult behavioral and metabolic interpretations.

The exercise section is interesting but overinterpreted. The conclusion that physical activity is insufficient as therapy for chronic BDNF deficiency is not justified from short-term moderate-intensity protocols alone. Different intensities, durations, ages of initiation, environmental enrichment, resistance training, or combined pharmacological approaches may produce different outcomes. The authors should narrow this conclusion to the specific interventions studied.

Minor Comments

The manuscript requires substantial stylistic editing. Several sentences are too long, repetitive, or syntactically awkward. Terms such as “knockout” should be used carefully, since the focal animals are heterozygous haploinsufficient rats rather than complete knockouts. “Gender” should be replaced by “sex” for animal studies.

The figures are conceptually useful but visually crowded. Figure 2 in particular presents a speculative mechanistic network with insufficient distinction between demonstrated findings and hypothesized links. The figure should use different graphical conventions for direct evidence, inferred pathways, and proposed mechanisms.

Table 1 is valuable but overly dense and should be reorganized. A more useful version would separate study design, sample characteristics, primary endpoint, principal finding, and translational relevance. It should also identify which studies used the SD-BDNFtm1sage heterozygous line as the primary model.

The manuscript would benefit from a short section on limitations of the review itself, including absence of systematic search methods, likely publication bias, small preclinical evidence base, and lack of formal quality assessment.

Comments on the Quality of English Language

Author Response

Author(s): Lucyna Mrówczyńska and Włodzimierz Mrówczyński

Title: Multisystemic consequences of Brain-Derived Neurotrophic Factor (BDNF) haploinsufficiency in the SD-BDNFtm1sage Rat Model

Response to Reviewer 2

Thank you for your insightful comments, which have significantly contributed to improving our manuscript. In accordance with your suggestions, we have revised the text accordingly; all modifications are highlighted in purple. Below, we provide detailed responses to the comments raised by Reviewer 2.

Point 1

In the second paragraph of the new Limitations and methodological considerations section, we have added text clarifying that Shirayama [84] and Yang [85] used wild‑type Sprague–Dawley rats, with homozygous knockout lines serving solely as antibody controls.

It should be noted that several of the mechanisms discussed here were not confirmed in the heterozygous rat line used in this study. Experiments conducted by Shirayama et al. [88] and Yang et al. [89] were conducted on wild-type Sprague–Dawley rats, and homozygous tissue with the Bdnf gene knocked out served only as an antibody control. Their results were included to broaden the interpretive context, but they do not demonstrate the presence of these mechanisms in SD-BDNFtm1sage heterozygotes. Table 1 provides a detailed summary of the model, genotype, and intervention for each of the cited studies.

The full classification of individual studies is presented in Table 1, which has been substantially revised to improve clarity.

Point 2

We agree with Reviewer 2 that the original wording regarding causality was overly categorical.

We have softened the wording “causes” in §3.4 replacing it by is “associated with”

These results indicate that reduced BDNF signaling is associated with parallel metabolic and cardiovascular pathology in conjunction with the neurological dysfunctions described above – arguably the most unexpected systemic consequence of BDNF haploinsufficiency in this model.

We have reworded the captions for Figure 2 to improve clarity and ensure accurate interpretation.

Figure 2. Self‑reinforcing corticolimbic–systemic feedback loop in SD-BDNFtm1sage rats. Prefrontal BDNF deficiency is proposed to drive HPA‑axis dysregulation, glucocorticoid excess, and REM‑sleep disruption, which may in turn exacerbate peripheral metabolic and cardiovascular pathology. The diagram reflects an integrative hypothesis rather than a proven causal pathway.

In the third paragraph of the new Limitations and methodological considerations section, we have added text presenting this cycle as a hypothesis that requires validation through longitudinal studies, tissue‑specific rescue approaches, TrkB modulation, endocrine profiling, and mediation analyses.

The feedback loop between the cortical-limbic and the systemic systems, proposed in Section 3 and illustrated in Figure 2, should be regarded as an integrative hypothesis rather than a proven causal pathway. Establishing causality will require specifically designed experiments, including longitudinal studies, tissue-specific or time-controlled rescue of TrkB modulation, endocrine profiling, and formal mediation analyses.

Point 3

In the first paragraph of the new “Limitations and methodological considerations” section, we added required explanations:

Several limitations of the underlying evidence base and of the present review should be acknowledged. The literature on the SD-BDNFtm1sage line is small and methodologically heterogeneous: the available studies differ in sample size, sex inclusion, age at testing, housing and husbandry, stress exposure, behavioural protocols, the timing of tissue collection, and assay methodology. Such heterogeneity is particularly consequential for BDNF, whose measured levels are highly sensitive to pre-analytical and contextual factors including circadian timing, recent physical activity, acute and chronic stress, diet, platelet activation, and sample handling. The convergence of results highlighted in this review should therefore be viewed as a qualitative phenomenon intended to inform hypothesis generation.

Point 4

In the Conclusions section, we have added a sentence explicitly stating that the clinical utility of this biomarker is not yet sufficient for diagnostic or prognostic use, particularly in humans.

This potential must, however, be read cautiously. The data establish biological validity and—in rats—analytical detectability, but not clinical utility: serum BDNF is strongly affected by platelet count, physical activity, circadian rhythm, and sampling method, and rat-tissue correlations do not confirm human relevance. Analytical accuracy, biological validity, and clinical utility must therefore be established independently before any clinical use.

Point 5

In the fourth‑to‑last paragraph of the new Limitations and Methodological Considerations section, we have added a subsection assessing the strength of translational evidence: strongest for nociception (supported by human studies), moderate for affective processes, sleep, and reward, and weakest for metabolic and cardiovascular processes.

It should also be emphasized that the certainty of translational inferences is not uniform across all studies included in this review. It is strongest in the case of nociception, where reduced pain sensitivity has been consistently demonstrated in heterozygous rats and in humans with WAGR deletions who express BDNF [87]. It is moderate for affective, sleep, and reward phenotypes, which are well characterized in rats but rely on analogies to corresponding states in humans. In contrast, it is weakest for findings regarding metabolism and the cardiovascular system, which come from a small number of studies that have not been independently verified in relation to human diseases.

Point 6

In the third‑to‑last paragraph of the new Limitations and Methodological Considerations section, we have added the corresponding explanation.

Finally, the phenotypes described in this review may have multiple underlying explanations. Because the deletion is constitutive, some behavioral and metabolic features in adult individuals may result from developmental programming or compensatory mechanisms rather than from a persistent BDNF deficiency in mature tissue (see Introduction and Section 3.3). Furthermore, increased body weight, altered body composition, fatty liver, reduced brain and pituitary mass, and possible neuroendocrine differences in this strain may confound behavioral measures dependent on locomotion, motivation, or body weight, although strain-specific effects in this rat strain cannot be ruled out. Therefore, considerable caution should be exercised before attributing a given behavioral observation or metabolic outcome solely to BDNF signaling.

Point 7

This revision was made in accordance with Reviewer 2’s suggestions at the end of §3.5.

However, it should be emphasized that this conclusion applies only to short-term, moderate-intensity endurance training programs. Programs that differ in intensity, duration, or age of initiation, as well as other methods (e.g., strength training, environmental enrichment, or physical exercise combined with pharmacological treatment), have not been studied in this context and may produce different results.

Minor Comments

An explanation of the term “knockout” has been added at the end of the Introduction

To maintain consistency with the original studies and standardize terminology, heterozygous SD-BDNFtm1sage rats are referred to as “knockout rats” in some subsequent sections of this review. This term is used exclusively in reference to BDNF haploinsufficiency, meaning that these animals possess only one functional allele of the Bdnf gene.

The term “gender” has been replaced with “sex” in several places within Section 3.2 to ensure accurate and discipline‑appropriate terminology.

Figure 2 has been completely redesigned in accordance with the reviewer’s suggestions.

Table 1 has been substantially revised in accordance with the Reviewer’s suggestions.

As suggested by Reviewer 2, we have prepared a new chapter entitled Limitations and Methodological Considerations. This section consolidates all relevant constraints of the reviewed studies, clarifies methodological caveats, and outlines the key factors that may influence the interpretation of the presented findings.

Limitations and methodological considerations

The feedback loop between the cortical-limbic and the systemic systems, proposed in section 3 and illustrated in Figure 2, should be regarded as an integrative hypothesis rather than a proven causal pathway. Establishing causality will require specifically designed experiments, including long-time studies, tissue-specific or time-controlled rescue of TrkB modulation, endocrine profiling, and formal analyses.

Finally, the phenotypes described in this review may have multiple underlying explanations. Because the deletion is constitutive, some behavioral and metabolic features in adult individuals may result from developmental programming or compensatory mechanisms rather than from a persistent BDNF deficiency in mature tissue (see Introduction and section 3.3). Furthermore, increased body weight, altered body composition, fatty liver, reduced brain and pituitary mass, and possible neuroendocrine differences in this strain may confound behavioral measures dependent on locomotion, motivation, or body weight, although strain-specific effects in this rat strain cannot be ruled out. Therefore, considerable caution should be exercised before attributing a given behavioral observation or metabolic outcome solely to BDNF signaling.

It should be also emphasized that although the SD-BDNFtm1 line is derived from the Sprague –Dawley strain with a low degree of inbreeding, it nevertheless exhibits some genetic heterogeneity, and the deliberate mutation does not rule out the possibility of unknown changes introduced during genome editing. Therefore, as with all genetically modified animals, this model also raises ethical concerns that require strict adherence to the 3Rs principle (replacement, reduction, refinement) and careful monitoring of animal welfare.

Additionally, a reduction of approximately 50% in this highly pleiotropic neurotrophin does not allow for a full understanding of the biology of BDNF and its role in development, cell-type-specific signaling, the proBDNF/mBDNF balance, or gene-environment interactions. For these reasons, the SD-BDNFtm1 line should therefore be viewed as a useful starting point for analysing BDNF-dependent mechanisms, rather than as a comprehensive substitute for human disease.

The heterozygous rat with BDNF deficiency described in this review shares several key features with mouse models of BDNF haploinsufficiency. In both groups, an approximately 50 percent decrease in BDNF levels is observed, along with impairments in cognitive and affective functions [69, 70, 75, 90]. However, there are two significant differences that limit the possibility of direct interspecies comparisons. First, mice exhibit marked hyperactivity [67], whereas rats are characterized by reduced motivation to explore [79]. Second, in mice, BDNF haploinsufficiency causes severe, late-onset obesity due to hyperphagia [95, 96], which can be reversed by BDNF administration [67]. Despite the analogous genetic alteration, this obesity phenotype has not been described in heterozygous rats. These discrepancies indicate that the behavioral and metabolic consequences of partial BDNF loss manifest differently in these two species. Therefore, caution should be exercised when generalizing the results of studies conducted in different models of BDNF deficient rodents.

Comments on the Quality of English Language

We appreciate the Reviewer’s observation regarding the need for substantial stylistic editing. The entire manuscript has undergone a thorough language revision to improve clarity, precision, and readability. Long or syntactically awkward sentences have been shortened or restructured, repetitive phrasing has been removed, and transitions between sections have been refined to ensure a smoother narrative flow.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

IJMS - ijms-4376775 Manuscript: Multisystemic consequences of Brain-Derived Neurotrophic Factor (BDNF) haploinsufficiency in the SD-BDNFtm1sage Rat Model

The narrative review authored by Mrówczyńsk and Mrówczyński examines the main research studies conducted on the SD-BDNFtm1sage rat model, a genetically modified rodent line. The article focuses primarily on studies that highlight the pleiotropic functions of this neurotrophin, thus providing a broad perspective and considering multiple viewpoints. Through the detailed and comprehensive “lens” of scientific reports focused on the consequences of BDNF deficit in mutant vs wild type rats, this review points out that BDNF directly participates in key brain-periphery interactions, involving vulnerability to mental disorders, disturbed sleep , metabolic and redox changes, altered nociception and neuromuscular abilities.

Furthermore, although these mutant rats have a lifespan similar to that of wild-type rats (Figure 1), they exhibit alterations in stress-dependent physiological circuits and networks, diminished explorative behaviors, accompanied by anatomical changes in brain regions such as the midbrain, the limbic system, and the cortex, while also revealing an impact on other non-neural vital organs such as liver. This genetic model thus provides further evidence of BDNF’s involvement in the defense and maintenance of homeostasis in various organs and tissues, in addition to the brain. The review also supports the idea that there may be preferential cellular targets for the genetic haploinsufficiency of Bdnf. This article is therefore original and promising for future investigations aimed at further clarifying the roles of BDNF, opening new avenues of research in the field of the pathophysiology of neuropsychiatric disorders and their link to somatic comorbidities.
It is noteworthy that the authors treat BDNF and its receptor as key biomolecules capable of stimulating and modulating flexible gene x environment adaptive cascades, based on the results obtained using this specific genetic model presenting only one functioning allele, which has been relatively overlooked in this regard. In summary, this rat model may help elucidate the pathophysiology of psychiatric disorders, even though the BDNF reduction apparently cannot yet be linked to distinct behavioral and metabolic phenotypes in psychiatric patients.

I suggest some revisions to further improve and refine the manuscript:

- The “Introduction” section should provide a more detailed description of this BDNF genetic defect model at the molecular level: it should include a brief overview of the structure of the BDNF gene in rats and its features compared to the human gene; furthermore, a short but more in-depth account should be provided of the genetic characteristics of BDNF in these specific mutant strains under study, along with how they were generated. The possible presence of other genetic markers besides BDNF in these same rat strains should also be discussed.

Moreover:
- Have any data or results concerning platelets and/or lymphomonocytes and endothelial cells ever been obtained in this model? Indeed, many peripheral investigations on BDNF have been conducted in plasma, serum, platelets and immune system cells (Serra-Millàs M. Are the changes in the peripheral brain-derived neurotrophic factor levels due to platelet activation? World J Psychiatry. 2016;6(1):84-101; Wang and Tian. Brain-derived neurotrophic factor in autoimmune inflammatory diseases. Exp Ther Med. 2021;22(5):1292). Serum BDNF corresponds more to alpha granule intraplatelet contents while plasma amounts of this neurotrophin are rather linked to brain, endothelial, platelet derivation, reflecting different pools of the neurotrophin in peripheral districts. Platelets are the main reservoir of BDNF in the body (Boukhatem et al. The intriguing role of platelets as custodians of brain-derived neurotrophic factor. Res Pract Thromb Haemost. 2024;8(3):102398; Bouhaddou et al. The link between BDNF and platelets in neurological disorders. Heliyon 2024;10(21):e39278); The authors should discuss also these aspects.

- Finally, the limitations of mutant genetic models such as this one should be acknowledged. These include for instance the use of inbred animal lines, ethical concerns, the possibility of unrecognized additional mutations not yet tested, and the limited ability of the model to fully reproduce the complexity of BDNF-related mechanisms and their contribution to human cognitive, behavioral, and affective disorders. A brief discussion of these issues would strengthen the manuscript.

Others: page 14. The sentence: “Importantly, BDNF reduction is not tissue-specific: significant deficits are observed in serum, prefrontal cortex, hippocampus, amygdala, and liver, while skeletal muscle and cardiac BDNF levels remain unchanged.” I suggest modifying the text for the sake of clarity, for instance as:” Importantly, BDNF reduction involves many tissues: significant deficits are observed in serum, prefrontal cortex, hippocampus, amygdala, and liver; anyway, skeletal muscle and cardiac BDNF levels remain unchanged.”

- The authors should adopt the nomenclature for rat genes, for instance Bdnf, within the manuscript.

Author Response

Author(s): Lucyna Mrówczyńska and Włodzimierz Mrówczyński

Title: Multisystemic consequences of Brain-Derived Neurotrophic Factor (BDNF) haploinsufficiency in the SD-BDNFtm1sage Rat Model

Response to Reviewer 3

Thank you for your valuable comments, which have helped us improve our article. In line with your suggestions, we have made the necessary changes to the text, which we have highlighted in green. Below are our responses to the comments made by Reviewer 3.

Point 1

I suggest some revisions to further improve and refine the manuscript:

As suggested by Reviewer 3, we have added information on the molecular structure of BDNF and the key differences between the rat and human genes at the first part of Chapter 2. In addition, following the recommendation of another reviewer, the description of the deletion procedure used to generate the SD‑BDNFtm1sage line has been added as second paragraph to improve methodological clarity.

At the molecular level, the rat Bdnf gene has a structure as complex as that of the human gene: numerous 5′ alternative non-coding exons, each with its own promoter, are spliced together with a single 3′ coding exon, which encodes the full-length preproBDNF protein. This conserved multi-promoter structure, combined with a high degree of coding sequence similarity, underscores the importance of BDNF research in rodents for translational studies. One significant difference between species is that the Val66Met polymorphism in humans (rs6265), which affects activity-dependent BDNF release and is associated with several neuropsychiatric traits, does not occur naturally in rats.

The SD-BDNFtm1Sage line was generated in a Sprague–Dawley rat genetic background by SAGE Labs using CompoZr zinc finger nuclease (ZFN) technology [71, 73]. A pair of ZFNs targets the Bdnf protein-coding region, inducing a double-strand break that is repaired by non-homologous end joining (NHEJ). This results in a small deletion and a premature stop codon, leading to the production of a truncated, non-functional protein instead of mature BDNF. This procedure is sufficient to disrupt all BDNF production, because although the rat Bdnf gene contains multiple alternative 5′ non-coding exons driven by separate promoters, all transcripts share a common 3′ coding exon [74]. Consequently, a mutation in this shared exon affects every transcript variant, regardless of the promoter used, preventing functional BDNF expression from the mutated allele. As a result, homozygous rodent mutants are completely depleted of BDNF and die within the first few days after birth, whereas heterozygotes retain one intact allele and survive lifelong with an approximately 50% reduction in BDNF expression [see Envigo technical summary of the BDNF‑KO rat model, which provides the most comprehensive overview of the homozygous and heterozygous phenotypes at: https://insights.envigo.com/hubfs/resources/data-sheets/envigo-gems-bdnf-ko-rat.pdf].

Point 2

Moreover: - Have any data or results concerning platelets and/or lymphomonocytes and endothelial cells ever been obtained in this model? Indeed, many peripheral investigations on BDNF have been conducted in plasma, serum, platelets and immune system cells (Serra-Millàs M. Are the changes in the peripheral brain-derived neurotrophic factor levels due to platelet activation? World J Psychiatry. 2016;6(1):84-101; Wang and Tian. Brain-derived neurotrophic factor in autoimmune inflammatory diseases. Exp Ther Med. 2021;22(5):1292). Serum BDNF corresponds more to alpha granule intraplatelet contents while plasma amounts of this neurotrophin are rather linked to brain, endothelial, platelet derivation, reflecting different pools of the neurotrophin in peripheral districts. Platelets are the main reservoir of BDNF in the body (Boukhatem et al. The intriguing role of platelets as custodians of brain-derived neurotrophic factor. Res Pract Thromb Haemost. 2024;8(3):102398; Bouhaddou et al. The link between BDNF and platelets in neurological disorders. Heliyon 2024;10(21):e39278); The authors should discuss also these aspects.

A corresponding paragraph has been added at the end of Section 2.2, as suggested by Reviewer 3, to address this important issue and ensure that the manuscript provides the necessary molecular and conceptual context for interpreting the findings.

Another issue concerns the cellular sources of circulating BDNF in these studies. Serum BDNF concentrations primarily reflect the large pool of this factor stored in the α-granules of platelets, whereas plasma contains only a small fraction of free BDNF derived from the brain, endothelium, and platelets [82]. Since platelets are the main reservoir of BDNF in the body [83, 84], serum concentrations are strongly influenced by platelet count and platelet activation, while immune cells and the endothelium provide additional BDNF, particularly in inflammatory conditions [85]. However, no studies have been conducted in the SD-BDNFtm1sage rats line specifically regarding BDNF in platelets, peripheral blood mononuclear cells, or the vascular endothelium. The reported decreases in serum concentrations [76] were measured using a total serum immunoassay without distinguishing between these compartments. Determining whether the deficiency reflects altered platelet content, reduced production by immune or endothelial cells, or changes in release would clarify the biological significance of BDNF changes in peripheral blood in this model and remains an important direction for future research.

Finally, the limitations of mutant genetic models such as this one should be acknowledged. These include for instance the use of inbred animal lines, ethical concerns, the possibility of unrecognized additional mutations not yet tested, and the limited ability of the model to fully reproduce the complexity of BDNF-related mechanisms and their contribution to human cognitive, behavioral, and affective disorders. A brief discussion of these issues would strengthen the manuscript.

As suggested by Reviewer 2 and 3, a new chapter entitled “Limitations and methodological considerations” has been prepared. At the end of this chapter, respective explanation concerning limitations of mutant genetic models has been added.

It should be also emphasized that although the SD-BDNFtm1 line is derived from the Sprague –Dawley strain with a low degree of inbreeding (which limits artifacts associated with variants found in fully inbred lines), it nevertheless exhibits some genetic heterogeneity, and the deliberate mutation does not rule out the possibility of unknown changes introduced during genome editing. Therefore, as with all genetically modified animals, this model also raises ethical concerns that require strict adherence to the 3Rs principle (replacement, reduction, refinement) and careful monitoring of animal welfare.

The sentence on page 14 has been revised according to the reviewer’s suggestion. The updated version now reads more clearly and distinguishes between tissues showing reduced BDNF levels and those in which BDNF remains unchanged.

Importantly, BDNF reduction involves many tissues: significant deficits are observed in serum, prefrontal cortex, hippocampus, amygdala, and liver; whereas skeletal muscle and cardiac BDNF levels remain unchanged

- The authors should adopt the nomenclature for rat genes, for instance Bdnf, within the manuscript.

It was done according to suggestion.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Please enhance the quality, clarity, and readability of the figures and tables throughout the manuscript.

Comments on the Quality of English Language

Author Response

Response to Reviewer 2

We thank the Reviewer for all comments. Suggested changes have been introduced to the revised version of manuscript.

Point 1

„The manuscript requires substantial stylistic editing. Several sentences are too long, repetitive, or syntactically awkward."

We have edited the text to improve its conciseness and readability. Excessively long sentences were shortened or split, redundant phrases were removed (e.g., “in the architecture of neural circuits, which in such an animal model…” was simplified to “in the architecture of neural circuits, which makes it difficult…”), and awkward constructions were reworded (e.g., “On the other hand, it has been shown that numerous stressors reduce…” → “On the other hand, numerous stressors reduce…”). One redundant concluding sentence in the Introduction and a repeated phrase (“in this rat line”) in the Limitations section were removed.

Point 2

„Terms such as 'knockout' should be used carefully, since the focal animals are heterozygous haploinsufficient rats rather than complete knockouts."

Throughout the text and in Table 1, the term “knockout” has been replaced with “heterozygous” or “haploinsufficient.” The term “knockout” has been retained only in reference to homozygous antibody control tissue and general “gene-knockout models.”

Point 3

“Gender' should be replaced by 'sex' for animal studies."

The word “gender” has been replaced by the word “sex” throughout the text.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The review ijms-4376775 written by Mrówczyńska and Mrówczyński and titled : "Multisystemic consequences of Brain-Derived Neurotrophic Factor (BDNF) haploinsufficiency in the SD-BDNFtm1sage Rat Model" has been carefully revised, also following all suggestions provided. The article can be now accepted for publication in IJMS. I suggest to the authors to verify reference section matching. It appears correct, but a rapid check is always necessary.

Author Response

Response to Reviewer 3

We thank the Reviewer for the positive assessment and for recommending acceptance.

Point 1

I suggest the authors verify reference section matching. It appears correct, but a rapid check is always necessary."

We verified all citations in the text against the reference list. All citations correspond to the correct references; every reference has been cited, and the references are numbered in the order of their first appearance.

We identified one duplicate: the paper by Lyons et al. (1999) was listed twice, as entries [66] and [95]. We combined them into a single entry [66], removed the duplicate, and updated the corresponding citation in the text; after renumbering, the list contains 95 bibliographic entries and has no gaps.

Author Response File: Author Response.docx