Review Reports - Contained Ensiling of High-Lipid Perennial Ryegrass: Fermentation Quality, Fatty Acid Retention, and Storage Stability

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,
I have reviewed the manuscript and provided comments for further improvement.
Please consider the comments.
Sincerely

L42-43: Remove keywords that are identical to those in the manuscript title, and limit the
number of keywords to 5
L48-49: The text correctly points out that GE does not translate directly to ME. However:
“resulting in approximately 1 MJ kg⁻¹ of more gross energy (GE)”. Suggestion:
emphasize from the beginning that this gain is potential, not guaranteed in ME, to avoid
a simplified reading by less specialized reviewers.
L58-61: The discussion about CH₄ is adequate and balanced, but could be slightly
improved:“Based on prediction equations and in vitro studies…”It could explicitly
mention that in vivo evidence is still scarce or nonexistent, reinforcing the justification
for the study.
L78-93: A sentence like: “This study has an intentional methodological scope”
can fly defensively. Suggestion: “Thus, this study focuses on methodological
development…” Maintains the content, but with a more assertive and less justificatory
tone.
L100-105: The sentence, "The study was designed to generate and characterize
genetically modified forage material..." could be slightly shortened, as a similar idea
appears in the Introduction.
L110: “Seedings” should be corrected to “seedlings”. The description of initial
fertilization can be condensed.
L149: Suggestion: reduce basic explanations about nitrate metabolism, keeping only the
essentials.
L178: Clarify whether the amount of added sugar was the same for HME and the control
group (it is assumed that it was, but it is worth specifying).
L213: The chemical analysis section is very long; it's worth summarizing by citing only
the methodology. Excessive detail in GC-MS (temperatures, times, voltages) can be
summarized as "as described by…".
L277: Suggestions: add the statistical model used. Adjustment: “Error bars and ranges
given are standard error of mean” use standardized form: standard error of the mean
(SEM).
L298: In the phrase: “due to more favorable growing conditions, including higher
temperatures…” it enters the interpretive field and should be softened. “…with higher
yields observed during the spring-summer period (P < 0.01).” The causal explanation can
be moved to the Discussion. Observe this type of wording throughout the results section.
In the results section, the authors should exclusively describe the results found, and any
interpretation should be allocated to the discussion section.
L320: The expression "slightly declined" is appropriate, but it could be quantified in the
first sentence for more detailed solutions. Example: "...decreased by approximately 0.5
percentage points of dry matter..."
L322: Avoid explanatory language in the results: “…maximize the FA content of the leaf”
this is methodological justification, already described in Methods. There is slight
redundancy in repeating WSC values that are already clearly presented in Figure 2.
L337: “Analysis of variation (AOV)” – the more common term is “analysis of variation
(ANOVA)”, unless the journal explicitly accepts AOV. Standardize “dry weight” to “dry
matter (DM)”, consistent with the rest of the manuscript.
L344: In the phrase: “As shown in Figure 3…” the historical reference [7] could be
omitted, as the result is from the current study. Avoid terms such as “asymptomatic rise”,
which are interpretive.
L362: Standardize % DM vs % DW (use only DM). In the phrase: “representing a 76%
increase” it is acceptable, but ideally the calculation should be confirmed in the
supplementary text or legend.
L374: Place the meaning of total massa in the table footer. Use ns or p-values explicitly,
but not both inconsistently. “The predicted ME here did not differ statistically” this phrase
should be outside the table, only in the text.
L385: In the phrase: “suggesting altered microbial fermentation dynamics”. it is an
interpretation and should be moved to the Discussion section.
L400: Be careful with overly long sentences, especially in paragraphs about the FA
profile.
the
L445: Comment: Although the manuscript makes it clear that cultivation was carried out
under containment conditions, in some passages there is a risk that the reader may
interpret
results
as
directly
transferable
to
commercial production
systems.Suggestion: Insert an explicit statement delimiting that: the results demonstrate
technical solutions and compositional stability, but do not fully simulate field production
systems. It is recommended to include a comparative sentence, decreasing which
responses could be greater or lesser under field conditions (e.g., biomass production,
WSC, FA).
Comment: The limitations of vessel volume are recognized, but could be explored more
analytically. Suggestion: Explicitly relate root limitations to: reduced nitrogen absorption;
potential impacts on lipid synthesis; expected physiological differences between HME
and null genotypes. If possible, we suggest future experimental approaches (e.g.,
fertigation, more frequent nutrient replenishment, hydroponic cultivation) to mitigate this
effect.
Comment: The addition of sugar, while methodologically justified, complicates the
interpretation of the absolute values of GE and ME. Suggestion: Highlight more clearly
that: the relative energy difference between genotypes remains valid; the absolute ME
values should not be interpreted as representative of commercial systems. Consider
including a sentence indicating that future silages without supplementation would be
important for practical validation. Comment: The CH₄ mitigation estimates are proposed
with caution, but can still be interpreted as a strong inference. Suggestion: Reinforce that:
the estimates are based on empirical models; the actual effects depend on ruminal
adaptation, intake, and diet composition. It is recommended to avoid terms such as
"potential reduction" without further qualification, replacing them with "model-based
estimate" or "theoretical mitigation potential".
Style Adjustments: Reduce long sentences in dense paragraphs (especially 4.4 and 4.7) to
improve flow. Avoid repeating phrases like "this study demonstrates". Citations: Consider
including at least one recent additional reference (within the last 5 years) related to: lipid
stability in silages; effects of dietary lipids on ruminal fermentation.
L562: Comment: The conclusion emphasizes the workflow predictions, but could more
clearly highlight what the study proved that was not previously demonstrated. Suggestion:
State explicitly that the study demonstrates, for the first time (or systematically), that:
multiple cycles of cultivation under containment; silage; and prolonged storage do not
compromise the lipid advantage of the HME genotype. This increases the perception of
scientific novelty.
Comment: The link to subsequent studies is present, but could be more assertive.
Suggestion: Insert a final sentence reinforcing that: the workflow enables robust in vivo
studies, but does not replace the need for direct animal evaluation.
Additional comments: In the tables, remove the bold formatting from the column; leave
it only in the item descriptions in the first row. The p-value format should be: p Value.
Tables and figures should be aligned with the manuscript text.

Author Response

Dear Authors,

I have reviewed the manuscript and provided comments for further improvement.

Please consider the comments.

Sincerely

We thank the reviewer for careful evaluation of the manuscript and for the constructive and insightful suggestions. We have revised the manuscript, accordingly (highlighted in blue, unless otherwise indicated), as detailed below. Line numbers refer to the revised manuscript.

L42-43: Remove keywords that are identical to those in the manuscript title, and limit the number of keywords to 5

Response: Thank you for this suggestion. The keyword list has been revised to remove redundancy with the title and reduced to five keywords, in line with the journal’s guidelines.

Lines 40-41: Keywords: genetically modified forage; high-metabolizable energy; silage fermentation; fatty acids; long-term storage

L48-49: The text correctly points out that GE does not translate directly to ME. However:

“resulting in approximately 1 MJ kg⁻¹ of more gross energy (GE)”. Suggestion:

emphasize from the beginning that this gain is potential, not guaranteed in ME, to avoid

a simplified reading by less specialized reviewers.

Response: We agree and have revised the text to clarify that the observed increase in gross energy represents a potential energetic advantage. We now explicitly state that conversion to metabolizable energy is not guaranteed and requires validation through in vivo animal studies, reducing the risk of overinterpretation by non-specialist readers.

Lines 45-49: Highlighted in yellow due to similar comment made by the Reviewer 2, “In perennial ryegrass (Lolium perenne L.), this trait was designed to increase potential metabolizable energy (ME) via increased lipid-derived gross energy (GE), rather than directly measuring ME, and results in foliar fatty acid (FA) content up to 80% higher than conventional cultivars, corresponding to approximately 1 MJ kg⁻¹ dry matter (DM) higher GE.”

Lines 63-65: Highlighted in yellow due to similar comment made by the Reviewer 2, “Translating increased GE into ME and validating any mitigation potential requires in vivo animal studies, as intake regulation, rumen adaptation, and digestibility can substantially influence outcomes [8, 9].”

L58-61: The discussion about CH₄ is adequate and balanced, but could be slightly

improved:“Based on prediction equations and in vitro studies…”It could explicitly

mention that in vivo evidence is still scarce or nonexistent, reinforcing the justification

for the study.

Response: This point has been incorporated.

Lines 57-63: Highlighted in yellow due to similar comment made by the Reviewer 2, “Dietary lipids are known to affect methanogenesis through multiple mechanisms, including shifts in rumen microbial populations and hydrogen utilization; however, these effects are dose-dependent and strongly influenced by diet composition and intake context [6]. Based on prediction equations and in vitro evidence, increased FA content in HME ryegrass has been hypothesized to reduce enteric CH₄ emissions; however, empirical in vivo validation of CH₄mitigation by HME ryegrass remains limited [7].”

L78-93: A sentence like: “This study has an intentional methodological scope”

can fly defensively. Suggestion: “Thus, this study focuses on methodological

development…” Maintains the content, but with a more assertive and less justificatory

tone.

Response: We have revised this section to adopt a more assertive tone.

Lines 82-84: “Thus, this study focuses on methodological development by establishing a reproducible, containment-compatible workflow for producing, ensiling, and storing GM HME ryegrass suitable for downstream animal-based evaluation, rather than testing agronomic or nutritional treatment effects.”

L100-105: The sentence, "The study was designed to generate and characterize

genetically modified forage material..." could be slightly shortened, as a similar idea

appears in the Introduction.

Response: The sentence has been deleted to avoid redundancy while essential information has been incorporated into Line 84 (see above response).

L110: “Seedings” should be corrected to “seedlings”. The description of initial

fertilization can be condensed.

Response: Line 114: This typographical error has been corrected.

The description of initial fertilization has also been streamlined to improve readability.

Lines 114-116: “After two weeks, the seedlings were supplied with a standard soluble fertilizer solution (Yates™ Thrive), consisting of 1 g/L at the rate of 200 mL/tray (30 x 45 cm2, 100 seedlings).”

L149: Suggestion: reduce basic explanations about nitrate metabolism, keeping only the essentials.

Response: We have deleted this general background information and retaining only details directly relevant to the experimental design and interpretation.

Lines 162-163: “ensuing uniform nitrogen availability and assimilation across genotypes during vegetative growth” was incorporated.

L178: Clarify whether the amount of added sugar was the same for HME and the control

group (it is assumed that it was, but it is worth specifying).

Response: This has been clarified in the Methods Section 2.6. Line 196, highlighted in yellow as similar comment to the Reviewer 2.

L213: The chemical analysis section is very long; it's worth summarizing by citing only

the methodology. Excessive detail in GC-MS (temperatures, times, voltages) can be

summarized as "as described by…".

Response: We agree and have substantially condensed this section. Detailed GC-MS parameters have been replaced with a reference to previously published methods, improving conciseness without loss of reproducibility.

Lines 277-281: “Volatile FA (VFAs) were extracted from 5 g of fresh silage in 50 mL of deionized water by blending to create a slurry, followed by incubation in a closed container with shaking at 130 rpm at room temperature for 1 h. The extract was filtered through a 100-μm mesh and stored at -80 °C until analysis. VFA concentrations were determined by GC-MS following established method described by Ghidotti et al [31].”

L277: Suggestions: add the statistical model used. Adjustment: “Error bars and ranges

given are standard error of mean” use standardized form: standard error of the mean

(SEM).

Response: The statistical model used has now been explicitly described. We have also standardized the terminology to “standard error of the mean (SEM)” throughout the manuscript.

Lines 299-301: “Data were analyzed using linear mixed-effects models implemented in R (version 4.5.1) using the lme4 and lmerTest packages and/or one-way analysis of variance (ANOVA), with genotype as the main effect.”

Lines 311, 330, 352, 371, 390, and 435: standard error of the mean.

L298: In the phrase: “due to more favorable growing conditions, including higher

temperatures…” it enters the interpretive field and should be softened. “…with higher

yields observed during the spring-summer period (P < 0.01).” The causal explanation can be moved to the Discussion. Observe this type of wording throughout the results section.

In the results section, the authors should exclusively describe the results found, and any interpretation should be allocated to the discussion section.

Response: We have revised the Results section to ensure it is purely descriptive. Causal and interpretive statements have been moved to the Discussion, and similar wording has been reviewed and corrected throughout the Results section.

In the Result Section, Lines 320-321: “Higher biomass accumulation was observed during the spring–summer period com-pared with the autumn–winter period (P < 0.01).”

Accordingly, In the Discussion Section, Lines 489-492: “Additionally, the higher biomass observed during the spring–summer period may reflect more favorable growing conditions, including higher temperatures and longer daylight hours; however, these factors were not directly tested in the present study [44].”

Poorter, H; Niklas, K.J.; Reich, P.B.; Oleksyn, J.; Poot, P.; Mommer, L. Biomass allocation to leaves, stems and roots: me-ta-analyses of interspecific variation and environmental control. New Phytol. 2012, 193(1), 30–50.

Reference has been included and reference list numbers have been edited accordingly.

In the Result Section, Lines 343-349: “Fresh herbage DM content—approximately 10–15% DM for both HME and null ryegrass—did not significantly differ between genotypes.

In this study, BC was numerically lower in fresh HME herbage (576 ± 23.2 mmoles/kg DM) compared to null herbage (586 ± 28.0 mmoles/kg DM).

Ryegrass herbage was harvested in the morning, specifically between 9-10:30 am. WSC content during this period was relatively low in autumn–winter, with greater accumulation of high-molecular weight–WSCs across both genotypes (Figure 2).”

In the Discussion Section Lines 453-459: “Unlike field-grown ryegrass, plants cultivated in controlled environments are subject to conditions that may affect key fermentation parameters such as DM, BC, and WSC levels. BC describes how resistant forage is to pH change, and forages with higher BC require more acid production (and thus more WSC) to achieve preservation pH [37]. Additionally, harvesting in the morning, especially during autumn–winter when sunrise is delayed, may limit WSC accumulation and influence fermentation potential [23].”

In the Result Section, Lines 404-411: “The acetic acid (HAc) content was significantly lower in HME silage (0.83% DM) compared to null silage (1.58% DM; P < 0.001) (Table 2). This value in HME silage falls slightly below the typical range for HAc concentrations in grass forage (1–3%) [37]. Butyric acid in both silages was within the expected range of 0.1–0.5% DM, and no other VFAs were detected in the silage of either genotype.

LA levels and pH did not differ significantly between genotypes. Other fermentation and compositional parameters, including BC and NH₄⁺–N, also showed no significant differences between the two genotypes (Table 2).”

In the Discussion Section Lines 572-584: “HME silage exhibited significantly lower HAc concentrations than null silage, while butyric acid remained within acceptable ranges for both genotypes [37]. These differences may reflect altered microbial fermentation dynamics associated with higher lipid content, as certain FAs can inhibit HAc-producing microorganisms [68]. Despite this difference, LA levels, pH and other key fermentation parameters remained stable across storage durations, indicating effective fermentation and long-term stability.

The advantageous characteristics of HME silage for at least a year storage highlights the practical value of the trait. The higher GE and ME content of HME silage is consistent with previous reports of increased energy density in HME ryegrass [1, 4] and supports its suitability for controlled animal evaluation. These findings suggest that the HME modification did not adversely impact silage quality or nutritive value. The observed improvements appear to be specifically associated with enhanced energy-related traits rather than broad alterations to fermentation stability or the nutritional profile.”

In the Result Section, Lines 413-433: “Storage duration influenced silage fermentation and composition (Table 3). DM content remained relatively stable over time and pH values across all storage periods stayed within the optimal range for both genotypes, consistently below 4.1.

HME silage maintained significantly higher FA levels than null silage at both 52 and 342 DAS, with a stable difference of 2.1–2.4% DM (Table 3). HME silage contained 5.75% FA compared to 3.59% in null silage; after 342 DAS, the difference was 5.17% vs. 2.80% DM. The FA loss between 52 and 342 DAS was statistically significant (P < 0.01).

FA composition was significantly affected by genotype and storage duration. At 52 DAS, HME silage had higher proportions of C18:1 (8.90% vs 0.72%) and C18:2 (21.0% vs 6.97%) and lower proportions of C16:0 (11.2% vs 14.8%) and C18:3 (55.6% vs 74.0%) com-pared with null silage. By 342 DAS, null silage showed increased C16:0, C16:1, and C18:0 and reduced C18:3, while HME silage maintained a relatively stable FA profile with enrichment of C18:1 and C18:2.

GE was significantly higher in HME compared to the null at both storage intervals (Table 3). DMD increased significantly in HME compared to null after prolonged storage, resulting in higher predicted ME for HME silage (10.8–11.6 MJ/kg DM) than null silage (10.5–10.8 MJ/kg DM; P < 0.05).

HAc levels were consistently lower in the HME silage across storage periods. BC declined over time in both silages, from ~775–838 mmol/kg DM at 52 days to ~404–417 mmol/kg DM after 342 days. LA content declined from 16.2% to 10.8% DM in null silage and from 16.9% to 11.7% DM in HME silage.”

In the Discussion Section Lines 535-554: “Extended storage influenced silage fermentation quality and nutrient composition [37]. The stable DM and pH values indicate that fermentation and preservation were successful across both genotypes. The higher GE and DMD in HME silage translated to in-creased predicted ME, highlighting the potential nutritional benefits of HME forage even after prolonged storage.

Consistently lower HAc levels in HME silage, alongside the observed decline in BC over time for both silages, indicate that HME modifications did not adversely affect fer-mentation stability.

The central objective of this study was to determine whether elevated lipid content in HME ryegrass could be retained following ensiling and extended storage under containment. The pronounced differences in FA composition, with HME silage enriched in un-saturated C18:1 and C18:2 and lower saturated FA proportions, suggest that HME modifications confer greater FA stability during storage, whereas polyunsaturated FAs in null silage were preferentially degraded or oxidized over time. Observed shifts in FA composition during storage were consistent with those reported for conventional ryegrass silage [58] and did not negate the lipid advantage of the HME genotype.

FA class distribution was characterized but not optimized, and no inference is made regarding the mitigation efficacy of specific FA classes. Although polyunsaturated and medium-chain FAs have been associated with CH₄ suppression [6, 59, 60], the present study was not designed to assess mitigation outcomes.”

L320: The expression "slightly declined" is appropriate, but it could be quantified in the

first sentence for more detailed solutions. Example: "...decreased by approximately 0.5

percentage points of dry matter..."

Response: This has been addressed by adding quantitative values (e.g., percentage point changes in DM) in the opening sentence of the relevant paragraph.

Lines 339-341: “However, herbage FA contents in HME plants decreased by approximately 0.5% of DM from the first to seventh harvest (Figure 1, gold dashed lines).”

L322: Avoid explanatory language in the results: “…maximize the FA content of the leaf”

this is methodological justification, already described in Methods. There is slight

redundancy in repeating WSC values that are already clearly presented in Figure 2.

Response: Methodological justification has been removed from the Results section, and redundant repetition of WSC values already presented in Figure 2 has been eliminated.

Lines 347-349: “Ryegrass herbage was harvested in the morning, specifically between 9-10:30 am. WSC content during this period was relatively low in autumn–winter, with greater accumulation of high-molecular weight–WSCs across both genotypes (Figure 2).”

L337: “Analysis of variation (AOV)” – the more common term is “analysis of variation

(ANOVA)”, unless the journal explicitly accepts AOV. Standardize “dry weight” to “dry

matter (DM)”, consistent with the rest of the manuscript.

Response: Terminology has been standardized to “analysis of variance (ANOVA)” and “dry matter (DM)” throughout the manuscript.

L344: In the phrase: “As shown in Figure 3…” the historical reference [7] could be

omitted, as the result is from the current study. Avoid terms such as “asymptomatic rise”, which are interpretive.

Response: We agree. The historical reference has been removed from this sentence.

“an asymptomatic rise” has been replaced with “a gradual increase” (Lines 363-364).

L362: Standardize % DM vs % DW (use only DM). In the phrase: “representing a 76%

increase” it is acceptable, but ideally the calculation should be confirmed in the

supplementary text or legend.

Response: All references to dry weight have been standardized to dry matter (DM) throughout the manuscript. The calculation underlying the reported 76% increase has been checked and confirmed (Lines 382-383, and 398-399).

L374: Place the meaning of total massa in the table footer. Use ns or p-values explicitly,

but not both inconsistently. “The predicted ME here did not differ statistically” this phrase should be outside the table, only in the text.

Response: The definition of “total mass” has been shown in the Table 2-footer (Lines 391-393).

Replaced “ns” with explicit p-values (Table 3).

“The predicted ME here did not differ statistically” has been removed and rewritten in the main text. (Lines 387-388).

L385: In the phrase: “suggesting altered microbial fermentation dynamics”. it is an

interpretation and should be moved to the Discussion section.

Response: This interpretation has been removed from the Results section and incorporated into the Discussion (see above response).

L400: Be careful with overly long sentences, especially in paragraphs about the FA

profile.

Response: We have carefully revised this section to shorten long sentences and improve readability, particularly in paragraphs describing fatty acid composition and storage-related changes (Lines 420-425).

L445: Comment: Although the manuscript makes it clear that cultivation was carried out under containment conditions, in some passages there is a risk that the reader may interpret results as directly transferable to commercial production systems. Suggestion: Insert an explicit statement delimiting that: the results demonstrate technical solutions and compositional stability, but do not fully simulate field production systems. It is recommended to include a comparative sentence, decreasing which responses could be greater or lesser under field conditions (e.g., biomass production, WSC, FA).

Response, Lines 476-482: “Nevertheless, it is important to note that these results demonstrate technical feasibility, fermentation performance, and compositional stability of HME ryegrass during ensiling and extended storage, they do not fully simulate commercial field production systems. Under field conditions, environmental variability may result in greater or lesser responses in traits such as biomass production, WSC accumulation, and FA content, which could in turn influence fermentation dynamics and nutritional outcomes [18, 37].”

Comment: The limitations of vessel volume are recognized, but could be explored more

analytically. Suggestion: Explicitly relate root limitations to: reduced nitrogen absorption; potential impacts on lipid synthesis; expected physiological differences between HME and null genotypes. If possible, we suggest future experimental approaches (e.g., fertigation, more frequent nutrient replenishment, hydroponic cultivation) to mitigate this effect.

Lines 495-502: “The confined vessel volume used in this study likely imposed limitations on root development, with potential downstream effects on nutrient acquisition, particularly nitrogen uptake [42, 43]. Restricted nitrogen availability can influence carbon–nitrogen balance and may constrain lipid biosynthesis, especially in genotypes with elevated carbon demand such as HME lines. In fact, FA accumulation in ryegrass has been shown to respond positively to nitrogen supply [45], given the increased metabolic investment in FA synthesis in HME plants.”

Lines 504-508: “Future studies could mitigate these constraints through optimized fertigation strategies, more frequent nutrient replenishment, or the use of soilless cultivation systems to better sustain nutrient availability and root function. Such approaches would allow clear-er separation of genotype-driven metabolic effects from artefacts associated with physical root confinement [46,47].”

References included and reference numbers have been edited accordingly.

Tuxun, A.; Xiang, Y., Shao, Y.; Son, J.E.; Yamada, M.; Yamada, S.; Tagawa, K.; Baiyin, B.; Yang, Q. Soilless cultivation: Precise nutrient provision and growth environment regulation under different substrates. Plants (Basel). 2025, 14(14), 2203.
Fussy A.; Papenbrock, J. An overview of soil and soilless cultivation techniques—chances, challenges and the neglected question of sustainability. Plants (Basel). 2022, 11(9), 1153.

Comment: The addition of sugar, while methodologically justified, complicates the

interpretation of the absolute values of GE and ME. Suggestion: Highlight more clearly

that: the relative energy difference between genotypes remains valid; the absolute ME

values should not be interpreted as representative of commercial systems. Consider

including a sentence indicating that future silages without supplementation would be

important for practical validation.

Lines 585-591: “However, sugar supplementation during ensiling across treatments may influence the absolute values of GE and predicted ME reported here. While this approach does not affect the validity of the relative energy differences observed between HME and null genotypes, the absolute ME values should not be interpreted as directly representative of commercial silage systems. Future studies evaluating unsupplemented silages under field-relevant conditions will be important to validate the practical impact of HME forage on energy supply.”

Comment: The CH₄ mitigation estimates are proposed with caution, but can still be interpreted as a strong inference. Suggestion: Reinforce that: the estimates are based on empirical models; the actual effects depend on ruminal adaptation, intake, and diet composition. It is recommended to avoid terms such as "potential reduction" without further qualification, replacing them with "model-based estimate" or "theoretical mitigation potential".

Lines 595-600: “Based on FA enrichment levels observed here, empirical prediction equations suggested a theoretical potential reduction in enteric CH4 emissions of approximately 10% [6]. These estimates are derived from empirical models and do not account for ruminal microbial adaptation, voluntary intake responses, or broader diet composition, all of which are known to influence in vivo methane emissions [71].”

References included and reference numbers have been edited accordingly.

Moraes, L.; Strathe, A.B.; Fadel, J.G.; Casper, D.P.; Kebreab, E. Prediction of enteric methane emissions from cattle. Glob. Chang. Biol. 2014, 20(7), 2140–2148.

Style Adjustments: Reduce long sentences in dense paragraphs (especially 4.4 and 4.7) to improve flow. Avoid repeating phrases like "this study demonstrates".

Response: We thank the reviewer for this comment. We have revised Section 4.4 and 4.7 to shorten overly long sentences and improve clarity and flow. In addition, repetitive phrases such as “this study demonstrates” have been minimized to enhance readability.

Citations: Consider including at least one recent additional reference (within the last 5 years) related to: lipid stability in silages; effects of dietary lipids on ruminal fermentation.

Response: Reference have been included.

Álvarez-Torres, J.N.; Ramírez-Bribiesca, J.E.; Bautista-Martínez, Y.; Crosby-Galván, M.M.; Granados-Rivera, L.D.; Ramí-rez-Mella, M.; Ruiz-González, A. Stability and effects of protected palmitic acid on in vitro rumen degradability and fermentation in lactating goats. Ferment. 2023, 10(2), 110.

L562: Comment: The conclusion emphasizes the workflow predictions, but could more clearly highlight what the study proved that was not previously demonstrated. Suggestion: State explicitly that the study demonstrates, for the first time (or systematically), that: multiple cycles of cultivation under containment; silage; and prolonged storage do not compromise the lipid advantage of the HME genotype. This increases the perception of scientific novelty.

Comment: The link to subsequent studies is present, but could be more assertive.

Suggestion: Insert a final sentence reinforcing that: the workflow enables robust in vivo

studies, but does not replace the need for direct animal evaluation.

Response: The Conclusions section has been revised to explicitly state that this study demonstrates, for the first time in a systematic manner, that multiple cycles of cultivation under containment, ensiling, and prolonged storage do not compromise the lipid advantage of the HME genotype.

A final sentence has been added to reinforce that while the workflow enables robust and compliant in vivo studies, it does not replace the need for direct animal evaluation.

Lines 632-641: “In conclusion, this study systematically demonstrates, for the first time, that multiple cycles of cultivation under contained conditions, ensiling, and prolonged storage do not compromise the lipid advantage of the HME ryegrass genotype. HME silage maintained enhanced FA content, energy density, and stable fermentation characteristics across storage periods, without adversely affecting key nutritional parameters. These findings confirm the robustness of HME traits under controlled production and storage conditions and highlight the potential for developing energy-dense forage for ruminant nutrition. While the workflow established here provides a reliable framework for investigating HME forage traits and silage performance, it has also been successfully applied in an in vivo feeding trial (results published separately), demonstrating practical applicability.”

Additional comments: In the tables, remove the bold formatting from the column; leave

it only in the item descriptions in the first row. The p-value format should be: p Value.

Tables and figures should be aligned with the manuscript text.

Response: All tables have been reformatted accordingly. Bold formatting is now restricted to item descriptors in the first row, p-values are presented consistently as p Value, and all tables and figures have been carefully aligned with the manuscript text.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Introduction

Lines 45–49: I invite the authors to clarify that HME is a trait designed to increase potential ME via increased lipid-derived GE, rather than a direct measurement of ME, to avoid conceptual ambiguity for readers less familiar with forage energy systems.

Lines 47–49: The quantitative increase is compelling. However, the physiological or biochemical basis for this increase (e.g., altered lipid biosynthesis or reduced turnover) is not mentioned. Add a short mechanistic phrase (e.g., “via altered lipid biosynthesis and storage pathways”) if this information is well established in the cited literature.

Lines 50–52: This is an important point that strengthens external validity. However, the term “agronomically relevant conditions” is broad. Consider specifying whether this refers to field-grown perennial systems, multiple defoliations, or seasonal variation, to reinforce credibility.

Lines 53–57: I invite the authors to note that lipid effects on methane are dose-dependent and context-specific, to pre-empt overgeneralization.

Lines 56–57: The use of “hypothesised” is appropriate and cautious. However, the reliance on prediction equations and in vitro data could be emphasized more strongly as a limitation. Strengthen this sentence by explicitly stating that empirical in vivo validation remains limited or absent for HME ryegrass.

Line 58: before talking about CH4, dietary fats have also an impact on digestibility. I invite the authors to add these references:

Microalgae supplementation improves goat milk composition and fatty acid profile: A meta-analysis and meta-regression. Archives Animal Breeding 68.1 (2025): 223-238. https://doi.org/10.5194/aab-68-223-2025

Effect of dietary fat source and concentration on feed intake, enteric methane, and milk production in dairy cows. Journal of Dairy Science 108.1 (2025): 553-567. https://doi.org/10.3168/jds.2024-25446

Materials & Methods

Lines 106–127: Clarify whether pot size was included explicitly as a fixed effect or absorbed within the greenhouse round random effect, to avoid ambiguity in the statistical model specification.

Lines 142–147: Consider clarifying whether irrigation volumes were validated against substrate moisture or plant water status, or whether schedules were empirically determined.

Lines 149–161: Line 159–161 notes declining fertilizer availability per plant over time.
Please clarify whether this decline was quantified or monitored via plant tissue N or yield, as this may influence later compositional results.

Lines 178–197: Clarify whether sugar addition could influence subsequent comparisons between HME and null material, particularly for fermentation end-products.

Discussion

Line 447–453: The conclusion that silage quality was “acceptable” is appropriate but could be strengthened by briefly referencing key indicators (e.g., pH, LA, VFA profile).

Line 456–458: The phrase “addresses a key translational barrier” is justified, but consider explicitly stating that few studies have progressed GM forage traits to animal evaluation under containment, if accurate, to reinforce novelty.

Line 491–495: The argument that elevated FA content represents a genuine genotypic difference is valid. Explicitly state that energy comparisons should be interpreted cautiously due to external carbohydrate input, particularly if GE/ME differences are highlighted later in Results.

Author Response

We thank Reviewer 2 for careful reading of the manuscript and for the constructive, insightful feedback, which has substantially improved the clarity of the manuscript. All changes have been incorporated into the revised manuscript and are highlighted in yellow for your review.

Introduction

Response: We agree and have clarified this point to avoid conceptual ambiguity. The text has been revised to explicitly state that HME ryegrass is engineered to increase gross energy density via lipid accumulation, thereby increasing the potential metabolizable energy available to the animal, rather than directly measuring ME.

Lines 45-49: “In perennial ryegrass (Lolium perenne L.), this trait was designed to increase potential metabolizable energy (ME) via increased lipid-derived gross energy (GE), rather than directly measuring ME, and results in foliar fatty acid (FA) content up to 80% higher than conventional cultivars, corresponding to approximately 1 MJ kg⁻¹ dry matter (DM) higher GE.”

Response: We have revised this sentence to briefly indicate that increased energy density arises from altered lipid biosynthesis and storage pathways, consistent with established literature on HME ryegrass. This provides mechanistic context without expanding beyond the scope of the current study.

Lines 49-52: “This increase in foliar lipid content is achieved by enhanced triacylglycerol biosynthesis via overexpression of diacylglycerol acyltransferase 1, together with reduced lipid turnover associated with stabilization of lipid droplets by an engineered cysteine-oleosin [4].”

Response: We have revised the text to specify that this refers to perennial ryegrass grown under repeated defoliation in greenhouse conditions designed to mimic managed pasture systems, thereby strengthening external validity and avoiding overly broad terminology.

Lines 52-55: “Field validation studies in the United States have demonstrated that HME ryegrass can maintain elevated leaf FA and GE under repeated defoliation and agronomic conditions representative of managed perennial pasture system [5], supporting its potential to improve forage energy density.”

Lines 53–57: I invite the authors to note that lipid effects on methane are dose-dependent and context-specific, to pre-empt overgeneralization.

Response: We agree and have added wording to explicitly acknowledge that lipid-mediated methane mitigation is dose-dependent and influenced by diet composition and intake context, helping to prevent overgeneralisation of mitigation potential.

Lines 57-60: “Dietary lipids are known to affect methanogenesis through multiple mechanisms, including shifts in rumen microbial populations and hydrogen utilization; however, these effects are dose-dependent and strongly influenced by diet composition and intake context [6].”

Response: We have strengthened this statement to clearly acknowledge that methane mitigation estimates are based on prediction equations and in vitro evidence, and that empirical in vivo validation for HME ryegrass remains limited. This reinforces appropriate caution in interpretation.

Lines 60-65: Based on prediction equations and in vitro evidence, increased FA content in HME ryegrass has been hypothesized to reduce enteric CH₄ emissions; however, empirical in vivo validation of CH₄ mitigation by HME ryegrass remains limited [7]. Translating increased GE into ME and validating any mitigation potential requires in vivo animal studies, as intake regulation, rumen adaptation, and digestibility can substantially influence outcomes [8,9].

Line 58: before talking about CH₄, dietary fats have also an impact on digestibility. I invite the authors to add these references:

Microalgae supplementation improves goat milk composition and fatty acid profile: A meta-analysis and meta-regression. Archives Animal Breeding 68.1 (2025): 223-238. https://doi.org/10.5194/aab-68-223-2025

Response: We thank the reviewer for these suggestions and have incorporated the recommended recent references (Lines 677-681, ref#8,9) to acknowledge that dietary fat can influence digestibility, intake, and fermentation dynamics, in addition to methane emissions.

Reference numbers have been edited in the following text accordingly.

Materials & Methods

Response: We have clarified that pot size was consistent within greenhouse rounds and that variation associated with pot size was therefore absorbed within the greenhouse round random effect in the statistical model. This clarification has been added to the Methods section.

Lines 131-133: “Pot size was consistent within each greenhouse round and was treated as a fixed factor in statistical analyses. Therefore, any variation associated with pot size is captured by the greenhouse round random effect in the model.”

Lines 142–147: Consider clarifying whether irrigation volumes were validated against substrate moisture or plant water status, or whether schedules were empirically determined.

Response: We have clarified that irrigation schedules were empirically determined to avoid visible water stress and maintain consistent growth, rather than being adjusted based on direct substrate moisture or plant water status measurements.

Lines 153-155: “Irrigation volumes were determined empirically to avoid visible water stress and maintain consistent plant growth, rather than being adjusted based on direct measurements of substrate moisture or plant water status.”

Lines 149–161: Line 159–161 notes declining fertilizer availability per plant over time.

Please clarify whether this decline was quantified or monitored via plant tissue N or yield, as this may influence later compositional results.

Response: We agree this is an important point and have clarified that fertiliser inputs were not adjusted on a per-plant basis over time, and that declining nutrient availability was assumed via foliar N measurements, representing a limitation that may contribute to later compositional variation.

Lines 163-170: “Fertilizer inputs were applied at a consistent rate per plant throughout the experiment and were not adjusted over time. Foliar total nitrogen content, measured at each harvest, declined progressively over subsequent harvests within a round (data not shown), indicating a gradual decrease in nutrient availability that may contribute to observed compositional changes. As root biomass accumulated over time, the amount of liquid fertilizer received per plant effectively decreased with successive applications, which likely contributed to reduced plant growth at later harvests.”

Lines 178–197: Clarify whether sugar addition could influence subsequent comparisons between HME and null material, particularly for fermentation end-products.

Response: We have added clarification noting that sugar was added uniformly across all silage treatments to standardise fermentation, and that while this may influence absolute fermentation end-products, it does not confound relative comparisons between HME and null material.

Lines 195-203: have been incorporated regarding response to the comment by the Reviewer 3.

“Table sugar was added at 50 g kg-1 wilted fresh weight to standardize fermentation in small-scale laboratory silos, including both HME and null material. This approach was used to minimize confounding effects arising from seasonal and developmental variability in endogenous water-soluble carbohydrate (WSC) concentrations of greenhouse-grown ryegrass. Accordingly, the ensiling system was designed to support comparative assessment of genotype effects under a controlled fermentation environment, rather than to simulate commercial ensiling practice. While this addition may affect the absolute concentrations of fermentation end-products, it was applied uniformly and therefore does not con-found comparisons between HME and null silage.”

Discussion

Line 447–453: The conclusion that silage quality was “acceptable” is appropriate but could be strengthened by briefly referencing key indicators (e.g., pH, LA, VFA profile).

Response: We have revised this section to explicitly reference pH, lactic acid concentration, and volatile fatty acid profiles as indicators supporting the conclusion that silage quality was acceptable.

Lines 459-453: “Silage quality was considered acceptable, as indicated by pH values within the target range, high LA concentrations, and VFA profiles consistent with successful fermentation. This quality was achieved consistently across genotypes and storage durations, indicating that containment-compatible ensiling protocols can preserve forage material suitable for subsequent animal-based studies.”

Response: We have added a sentence explicitly stating that few genetically modified forage traits have progressed to animal evaluation under containment, comparing to evaluation studies in GM crops, reinforcing the translational significance and novelty of the study.

Lines 468-475: “Few studies have progressed genetically modified forage traits beyond molecular or agronomic assessment to controlled animal evaluation, particularly under containment conditions [2]. Although GM crops such as herbicide tolerant maize and soybean have been evaluated in animal feeding studies for safety and nutritional equivalence, the literature on transgenic forage plants evaluated in livestock feeding experiments remains limited [2, 41]. This relative paucity of controlled trials reinforces the translational significance and novelty of the present study.”

Raman, R. The impact of Genetically Modified (GM) crops in modern agriculture: A review. GM Crops Food. 2017, 8(4), 195–208.

Reference numbers have been edited in the following text accordingly.

Response: We agree and have explicitly stated that energy comparisons should be interpreted cautiously, particularly where gross or metabolizable energy differences are discussed, due to the addition of external carbohydrate during ensiling.

Lines 527-529: “Comparisons of energy content should be interpreted with caution, particularly where GE or ME are discussed, as the addition of external sugar during ensiling may influence absolute energy values.”

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Title: A Contained Workflow for Producing and Ensiling High-Lipid Genetically Modified Ryegrass for Downstream Animal Studies

Overview & General Comments

This manuscript describes a carefully executed workflow for the contained production, harvesting, and ensiling of genetically modified high-ME perennial ryegrass. It addresses a real bottleneck in GM forage research, especially in regions where regulatory constraints may limit filed-scale production. This study is well written, detailed, and acknowledges realistic limitations. I do have some questions, comments, and suggestions; please see the following major and minor comments.

Major Comments

This manuscript states that repeated harvests from the same plants are not independent biological replicates (L100 – L102) and that silos constitute the experimental unit for silage analyses (L102 – L103). This becomes somewhat less clear in the presentation and interpretation of results, especially in Table 3 (L433), where means +/- SEM are presented without clear indication of what constitutes n for each comparison. Please clarify.

The addition of table sugar (L187) to standardize fermentation is an understandable methodological choice under containment conditions. It is acknowledged that the addition of sugar alters the fermentation environment and may complicate interpretation of genotype-related differences (L542 – L544). I would consider introducing this information earlier (potentially in the Materials and Methods section) so that results are more readily interpreted as genotype effects expressed under a controlled fermentation system, rather than extrapolated to supplemented ensiling systems.

Based on information from Table 2 (L375 – L377), the rationale for selecting 12 silage batches from 56 produced should be clarified. If possible, please indicate how batches were chosen and whether selected batches were balanced (e.g. across greenhouse, season, storage duration, etc.).

I might have missed this in the Materials and Methods, and if I did, please ignore this comment. But it appears (aside from the Abstract) that the first mention of long-term storage analysis at 52 days and 342 days post-ensiling is first mentioned at L402 – L403 in the Results. Please include this time frame somewhere within your Materials and Methods. I have a similar comment related to Table 2. I did not see the information about batches until L366 and L374 - L377. Please clarify in the Materials and Methods.

The comparison of silage composition at 52 days vs. 342 days (Table 3) provides useful information on long-term stability. However, some statements within the Discussion imply mechanistic stability or preferential preservation of fatty acid classes (L504 – L506). This may hold true to a point, but I’m not sure this can be fully supported given the number of time points assessed. I would consider briefly mentioning this limitation here with backing references (if available) as well as where it was mentioned in L544 – L545.

Minor Comments

Please be consistent with the use of FA vs. FAs (ex. of FAs: L221, L222, L414, L420, L511, L522)

Author Response

Overview & General Comments

We thank the reviewer for careful reading of the manuscript and for constructive and insightful comments. We have addressed all points raised and believe the revised manuscript (highlighted in green) have improved clarity, transparency, and interpretation of the workflow and results. Detailed responses are provided below.

Major Comments

This manuscript states that repeated harvests from the same plants are not independent biological replicates (L100 – L102) and that silos constitute the experimental unit for silage analyses (L102 – L103). This becomes somewhat less clear in the presentation and interpretation of results, especially in Table 3 (L433), where means +/- SEM are presented without clear indication of what constitutes n for each comparison. Please clarify.

Response: We agree that this required clearer presentation. We have now explicitly clarified replication at both the Methods and table legend levels.

In Materials and Methods Section, lines 104-109: “Plants grown in individual pots constituted the biological experimental units for plant growth and herbage composition measurements. Repeated harvests from the same plants were treated as temporal observations and not as independent biological replicates. For all silage-related analyses, individual laboratory silos constituted the experimental units. Harvested material from multiple plants within a harvest was homogenized prior to ensiling, and each silo was treated as an independent analytical replicate.”

In Table 3 legend, lines 449-454: “Data represent means ± standard error of the mean calculated at the silo level, with each silo constituting one experimental unit. n= 6 independent silos.” (details have been provided in the Materials and Methods)

The limited number of silos selected for detailed long-term analysis is now explicitly acknowledged as a study limitation in the Discussion (Lines 555–560).

“Over the two assessed storage durations, while these observations describe trends in fatty acid composition, the underlying mechanisms cannot be determined from only two time points. We acknowledge that more frequent sampling over storage would be required to draw mechanistic conclusions. Such changes in plant FA composition can vary with storage conditions and duration, and observed trends in individual FA classes from only two times points cannot be mechanistically definitive [61].”

The addition of table sugar (L187) to standardize fermentation is an understandable methodological choice under containment conditions. It is acknowledged that the addition of sugar alters the fermentation environment and may complicate interpretation of genotype-related differences (L542 – L544). I would consider introducing this information earlier (potentially in the Materials and Methods section) so that results are more readily interpreted as genotype effects expressed under a controlled fermentation system, rather than extrapolated to supplemented ensiling systems.

Response: We agree and appreciate this suggestion. The rationale for sugar supplementation has now been introduced earlier in the Materials and Methods, allowing results to be interpreted explicitly as genotype effects expressed under a controlled fermentation system.

In the Materials and Methods Section, lines 196-201: “This approach was used to minimize confounding effects arising from seasonal and developmental variability in endogenous water-soluble carbohydrate (WSC) concentrations of greenhouse-grown ryegrass. Accordingly, the ensiling system was designed to support comparative assessment of genotype effects under a controlled fermentation environment, rather than to simulate commercial ensiling practice.”

In the Discussion Section, lines 609-612: As described in Section 2.6, sugar was added to standardize fermentation in small-scale silos under containment conditions. While this approach alters the fermentation environment relative to commercial systems, it may partially obscure intrinsic genotype-related differences in silage composition to be evaluated [74, 75].

Based on information from Table 2 (L375 – L377), the rationale for selecting 12 silage batches from 56 produced should be clarified. If possible, please indicate how batches were chosen and whether selected batches were balanced (e.g. across greenhouse, season, storage duration, etc.).

Response: Thank you for highlighting this point. We have clarified the batch selection strategy and its rationale.

Lines 224-237: “A total of 56 silage batches were produced (28 HME and 28 null).

For general chemical analysis of silage (Table 2), subsamples from 12 HME silage batches and 12 null silage batches were randomly selected (8 from Glasshouse A and 4 from Glasshouse B) and sampled between 50-80 days after ensiling (DAS). Independent selection for this analysis did not account for sequential harvests or seasonal effects and was intended to provide a representative snapshot of silage quality across containment facilities.

For long-term storage stability analyses (Table 3), silos were randomly selected across greenhouse facilities, sequential harvests, and seasons. To maintain paired comparisons, when an HME silo was selected, the corresponding null-genotype silo produced on the same day and under identical ensiling conditions was selected as its matched pair. Each selected silo was subsampled at 52 DAS, re-vacuum sealed, and subsequently resampled at 342 DAS, such that measurements at both time points were obtained from the same silo batches.”

I might have missed this in the Materials and Methods, and if I did, please ignore this comment. But it appears (aside from the Abstract) that the first mention of long-term storage analysis at 52 days and 342 days post-ensiling is first mentioned at L402 – L403 in the Results. Please include this time frame somewhere within your Materials and Methods. I have a similar comment related to Table 2. I did not see the information about batches until L366 and L374 - L377. Please clarify in the Materials and Methods.

Response: We agree and apologise for this oversight. We have revised Sections 2.6 (Ensiling Processes) and 2.7 (Chemical Analyses) to explicitly define the silage storage durations analysed. Details are now clearly described in the Materials and Methods prior to their presentation in the Results (same as above).

In Section 2.6, lines 224-237:

“A total of 56 silage batches were produced (28 HME and 28 null).

For long-term storage stability analyses (Table 3), silos were randomly selected across greenhouse facilities, sequential harvests, and seasons. To maintain paired comparisons, when an HME silo was selected, the corresponding null silo produced on the same day and under identical ensiling conditions was selected as its matched pair. Each selected silo was subsampled at 52 DAS, re-vacuum sealed, and subsequently resampled at 342 DAS, such that measurements at both time points were obtained from the same silo batches.”

In Section 2.7, lines 240-241: “Chemical analyses were conducted on silage material sampled at defined storage intervals, as described in Section 2.6, with each silo representing one analytical replicate.”

The comparison of silage composition at 52 days vs. 342 days (Table 3) provides useful information on long-term stability. However, some statements within the Discussion imply mechanistic stability or preferential preservation of fatty acid classes (L504 – L506). This may hold true to a point, but I’m not sure this can be fully supported given the number of time points assessed. I would consider briefly mentioning this limitation here with backing references (if available) as well as where it was mentioned in L544 – L545.

Response: We appreciate this important point and agree that mechanistic conclusions should be framed cautiously.

In Discussion Section, Lines 555-560: “Over the two assessed storage durations, while these observations describe trends in fatty acid composition, the underlying mechanisms cannot be determined from only two time points. We acknowledge that more frequent sampling over storage would be required to draw mechanistic conclusions. Such changes in plant FA composition can vary with storage conditions and duration, and observed trends in individual FA classes from only two times points cannot be mechanistically definitive [61].”

Lines 616-619: “It is also important to note that the current analysis was conducted on a subset of 12 (6 HME and 6 null) silage batches out of total 56 batches, which may limit the generalizability of the results, given the known variability in silage composition across batches and storage conditions [76].”

Abdiani, N.; Kolahi, M.; Javaheriyan, M.; Sabaeian, M. Effect of storage conditions on nutritional value, oil content, and oil composition of sesame seeds. J. Agric. Food Res. 2024, 16, 101117.

Reference has been included and reference list numbers have been edited accordingly.

Minor Comments

Please be consistent with the use of FA vs. FAs (ex. of FAs: L221, L222, L414, L420, L511, L522)

Response: We have standardised the terminology throughout the manuscript such that “FA” is used when referring to total fatty acid content collectively, while “FAs” is used only when referring to multiple individual fatty acid species (classes). All instances highlighted by the reviewer (e.g. Lines 248, 249, 530, 547, 558, 559, and 575) have been corrected accordingly.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I am convinced by the responses of the authors. For me, the paper is suitable for publication in its present form.

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for your thoughtful responses and edits to all comments and questions. At this time, I am satisfied with all changes and have no further comments or questions.