Impact of different mating and surgical protocols on the establishment of a mouse model for fetal scarless skin healing

Abstract

Background

The mouse fetal intrauterine wound healing model is crucial and commonly used for investigating mechanisms and evaluating potential therapies for scarless skin regeneration compared to fibrotic healing. However, traditional intrauterine surgery remains technically challenging and understudied, which is associated with high maternal mortality and pregnancy loss, prompting us to refine the surgical protocol. Here, we report how the choice of surgical and mating procedure impact outcomes obtained.

Methods

Pregnant mice underwent fetal surgery at embryonic days 15.5, 16.5 (E15.5, E16.5, scarless) and 18.5 (e18.5, fibrotic). Two surgical protocols were used: traditional method involved purse-string sutures, microsurgical scissors, amniotic fluid supplementation, and suture closure (Traditional); and our modified method omitting purse-string sutures, replacing scissors with needle puncture for uterine and fetal incisions, eliminating amniotic fluid supplementation, and employing skin staples for abdominal closure (Modified).

Results

The modified protocol significantly increased the likelihood of successful pregnancy, reduced operative time, decreased abortion rates, and enabled earlier modeling compared to the traditional method. At 48 h, 7 days, and 9 days post-surgery, E15.5 wounds healed scarlessly, displaying regenerated hair follicles and organized collagen. Conversely, E18.5 wounds formed typical fibrotic scars, characterized by dense, disorganized collagen without hair follicles.

Conclusion

The optimized surgical protocol presented here provides a simplified, reliable fetal mouse model with improved pregnancy success, reduced fetal loss, earlier implementation, and consistent phenotypic outcomes. This refined model enhances experimental efficiency, reproducibility, and animal welfare, having a major impact on mechanistic studies and therapeutic exploration for scarless skin regeneration.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12575-025-00290-5.

Introduction

Wound healing in adult skin is a complex, multi-phase process involving hemostasis, inflammation, proliferation, and remodeling. This often leads to scar formation characterized by dense, structurally disorganized collagen deposition and the loss of normal dermal appendages such as hair follicles [1]. Scar tissue typically exhibits four characteristic features: (1) dermal fibrosis, (2) loss of normal skin texture, (3) absence of appendages, and (4) altered pigmentation or tone [2]. Although scarring is a natural repair mechanism, can fulfill the requirements of restoring the skin barrier and preventing infection or dehydration, it results in tissue that is structurally and functionally inferior to uninjured skin, with reduced elasticity and barrier integrity. Beyond aesthetic concerns, these changes can impair quality of life and remain a significant clinical challenge across a variety of cutaneous injuries [3].

By contrast, mammalian fetal skin has a remarkable capacity for scarless regeneration before a certain gestational age. In mice, this regenerative ability is observed before embryonic day 16 (E16); in fetal lambs, wounds created before 120 days of gestation were shown to heal without scarring as early as 1971; and in humans, this phenomenon occurs before approximately 24 weeks of gestation [4–6]. Fetal skin injuries within this timeframe heal rapidly, with minimal inflammation and restoration of normal dermal architecture, including hair follicles and an organized collagen deposition as uninjured skin. However, this regenerative capacity is lost as gestation progresses; after E18 in mice, cutaneous wounds begin to heal with typical fibrotic scarring [5].

To understand the mechanisms underlying scarless versus fibrotic healing, and to identify potential therapeutic targets for adult scarless wound regeneration, fetal wound models based on intrauterine surgery have been extensively developed over the past three decades. Since the 1990 s, fetal wound models have shifted from large animal such as monkey and lamb to small rodent models, due to their short gestation periods, genetic tractability, and availability of molecular tools [7]. In mice, after multiple attempts [8–10], fetal mouse wound model have been established for direct and precise comparisons of scarless regenerative and fibrotic healing by performing intrauterine surgery at defined gestational ages—most notably at E16.5 (scarless) and E18.5/E19.5 (fibrotic) [7, 11, 12]. This fetal wound model enables direct comparison to identify molecular drivers of skin regeneration and essential modulators of fibrosis [11, 13]. Moreover, exposing E16.5 wounds to candidate pro-fibrotic mediators can reveal their capacity to induce scarring in an otherwise regenerative context.

Despite their utility, traditional intrauterine surgical protocols are technically challenging and associated with high rates of maternal mortality and fetal loss [7, 14]. These procedures typically include abdominal laparotomy, exteriorization of the uterus, precise placement of purse-string sutures on the uterine wall, uterine incision with microsurgical scissors, wound creation on the fetal dorsum, and amniotic fluid supplementation followed by layered closure with sutures [7, 14]. While effective in generating the desired scarless or fibrotic phenotypes, the technical complexity of the procedure often limits reproducibility, increases surgical time, and elevates the risk of fetal resorption, miscarriage, and maternal death.

Furthermore, variability in fetal number and pregnancy timing poses an additional challenge to implementing this model. Primiparous female mice often fail to produce sufficient litter numbers or become pregnant promptly through natural mating alone. The resulting limited litters and unpredictable pregnancy timing complicate experimental planning and reduce efficiency.

Additionally, most existing studies using the traditional method have successfully performed fetal wounding at E16.5 [7, 11–13]. However, because fetal development progresses rapidly, creating the wounding by even one day earlier significantly increases surgical difficulty. Previous studies have shown that the regenerative capacity vary on a daily basis [15], with earlier wounding offering greater potential for mechanistic discovery and therapeutic development.

Given these limitations, we aimed to develop a simplified, reproducible fetal wound healing model that preserves the essential phenotypic outcomes of the traditional method while improving animal welfare, procedural efficiency, and the possibility to create the fetal wound earlier. In this study, we present a refined intrauterine surgical approach that eliminates purse-string suturing, replaces scissor incisions with 19G needle puncture, omits amniotic fluid supplementation, and uses skin staples for closure. We further optimized mating protocols using hormone-induced superovulation to improve pregnancy success and fetal yield.

We hypothesized that our modified protocol would simplify the procedure and reduce operative time, thereby improving maternal survival, minimizing fetal loss, allowing earlier wounding time, and enhancing both experimental efficiency and consistency, while preserving the key phenotypic outcomes of scarless regeneration and fibrotic healing in fetal mice. By directly comparing surgical outcomes between the traditional and modified approaches, this study provides a valuable update to the field and offers a more accessible, efficient, and humane model for investigating fetal skin regeneration.

Methods

Animals and housing

All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Shanghai Jiao Tong University School of Medicine. C57BL/6 J male and female mice (6–20 weeks old) were purchased from GemPharmatech and housed under standard specific pathogen-free conditions (12 h light–dark cycle, 20–25 °C, 50–60% humidity) with free access to food and water.

Mating protocols and pregnancy confirmation

To ensure consistent embryo availability for intrauterine surgery, we compared two mating strategies:

Natural Mating (Traditional Method): Female and male mice were co-housed for 3 h (12:00–15:00), and vaginal plugs were checked the following morning (6:00–9:00). The day a vaginal plug was observed was recorded as embryonic day 0.5 (E0.5). Plug-positive females were separated and monitored for signs of pregnancy (abdominal enlargement with a characteristic diamond-shaped contour, increased body weight, uterine palpation revealing embryos with a soybean-like texture) beginning on embryonic day 7 (E7.5).

Hormone-Induced Superovulation (Modified Method): Female mice received an intraperitoneal injection of pregnant mare serum gonadotropin (PMSG, 5 IU) followed by human chorionic gonadotropin (hCG, 5 IU) 48 h later. Mating was conducted immediately after hCG injection. Vaginal plugs were monitored as above. Pregnancies were tracked based on plug detection (designated as embryonic day 0.5, E0.5), and surgeries were performed at E15.5, E16.5, and E18.5.

Surgical preparation

Before surgery, the operating table and all instruments were disinfected with 75% ethanol. All surgical tools and sterile packs containing gauze, cotton swabs, disposable surgical staples, and other required materials were prepared. The electric heating blanket was preheated in advance. Table 1 summarizes the supplies required for the procedure.

Table 1.

Supplies used in the surgery

Surgical instruments	Disposable sterile supplies	Medication	Others
Surgical scissors	Sterile cotton swab	Isoflurane (RWD)	Medical tape
Micro surgical scissors	Gauze	Depilatory cream
Tissue forceps	Syringe (1 ml, 20 ml)	Phosphate-buffered saline (sterilized)
Dressing forceps	Skin stapler	Iodine
Surgical eye forceps	19G needle	75% ethanol
Micro needle holder		Indian ink
Micro injector		95% saline

Surgical protocols

Fetal intrauterine surgery was performed under two protocols: a Traditional method based on prior literature [7, 11–13] and a Modified method developed in this study. All surgeries were conducted on E15.5 (modified protocol for scarless healing), E16.5 (traditional protocol for scarless healing) or E18.5 (modified and traditional protocol for fibrotic healing).

During the procedure, pregnant mice were anesthetized with 3–4% isoflurane in oxygen at 1 L/min for induction, followed by 1–1.5% isoflurane at 0.5 L/min for maintenance. Mice were placed in a supine position on a preheated electric heating pad. Deep tendon reflexes were assessed to confirm an adequate depth of anesthesia.

• Step 1: Uterine Exposure (consistency between traditional and modified protocol)

The abdominal region was depilated using hair removal cream, disinfected sequentially with iodine and 75% ethanol, and laparotomy was performed to expose the uterus. Warm sterile PBS was used to maintain tissue hydration and visibility during the procedure.

• Step 2: Fetal Skin Wounding (discrepancy between traditional and modified protocol)

Traditional protocol

A purse-string suture was placed on the uterine wall surrounding a selected fetus. The center of the suture loop was incised using microsurgical scissors to open the uterus. The amniotic sac was incised, and the fetal dorsum was positioned for skin excision. A 1-mm full-thickness dorsal skin wound was created on the fetus using scissors, and India ink was injected into the wound area for later identification. The uterine opening was closed by tightening the purse-string suture. Amniotic fluid was supplemented with prewarmed sterile saline.

Modified protocol

Select a fetus located near the uterine horn for optimal access, ensuring that its dorsal surface is clearly visible. Without using a purse-string suture, the selected fetus was manually stabilized via palpation. A 19G needle was used to puncture the uterine wall and amniotic sac in one step, avoiding major blood vessels (visible needle tip), then create a full-thickness 1-mm skin wound in the fetal dorsum (needle tip not visible). India ink was injected through the needle into the wound site for future localization. Slowly withdraw the needle and gently return the uterus into abdominal cavity, minimizing amniotic fluid leakage. No amniotic fluid supplementation was performed.

• Step 3: Maternal Laparotomy Closure (consistency between traditional and modified protocol)

For both protocols, the abdominal wall and skin were closed simultaneously using a surgical skin stapler. The surgical area was disinfected again, and the mother was placed in lateral recumbency on the heating blanket and monitored until full recovery of consciousness and mobility (typically within 30 min). Once fully mobile, the mouse was returned to its cage with free access to food and water.

Fetal collection and tissue processing

48 h post-surgery, pregnant mice were euthanized via CO₂ inhalation. The uterus was carefully excised, and fetuses were harvested. Gross morphology was assessed for wound appearance. India ink markings were used to locate wound sites on fetal dorsal skin.

For histological analysis, whole fetuses were fixed in 4% paraformaldehyde overnight, embedded in paraffin, and sectioned transversely.

Hematoxylin and Eosin (H&E) staining

Sections of the fetus were made at 10 μm and stained with H&E to evaluate tissue architecture. All slides were examined using NanoZoomer 2.0-HT (Hamamatsu Photonics, Hamamatsu, Japan). The number of the hair follicle was measured using ImageJ software version 1.53t (NIH, Bethesda, MD; https://imagej.nih.gov/ij, last access date: 2025.05.28). The quantifications were performed by individuals blinded to the experimental conditions.

Masson's trichrome staining

Sections of the fetus were made at 10 μm and stained with Masson's trichrome staining to evaluate collagen organization. All slides were examined using NanoZoomer 2.0-HT (Hamamatsu Photonics, Hamamatsu, Japan).

Outcome measures

The following parameters were recorded: Fetal number per pregnancy; surgical duration (from laparotomy start to skin closure); abortion rate: number of dead fetuses/total number of operated fetuses; post-injury wound phenotype: gross appearance at 2,7, and 9 days post-wounding; histological features: epidermal closure, dermal structure, presence of hair follicles, and collagen pattern.

Statistical analysis

Data were analyzed using GraphPad Prism (https://www.graphpad.com/scientific-software/prism/, last access: 2025.05.28). Quantitative comparisons between the two groups were conducted using the Student T test. For the quantitative figures, each dot represents one mouse. P < 0.05 was considered statistically significant. All analyses were performed with investigators blinded to the treatment group. In the figures, statistical significance is indicated as follows: * for P < 0.05, ** for P < 0.01, *** for P < 0.001, and **** for P < 0.0001.

Results

To validate the efficacy of modified model, fetal surgeries were performed at E15.5 (using modified protocol, Supplementary Video 1), E16.5 (using traditional protocol [16]) and E18.5 (using both modified and traditional protocol) in this study.

Comparison of traditional and modified surgical procedures

To visualize the differences between traditional and modified surgical approaches, we first compared procedural steps in both protocols (Fig. 1). Uterine exposure (Step 1) was consistent in both methods, involving anesthesia, hair removal, disinfection (Fig. 1A), and midline laparotomy to expose the uterus (Fig. 1B).

Fig. 1.

Comparison of traditional and modified intrauterine fetal wound surgery procedures. A Abdominal hair removal using depilatory cream prior to disinfection. Disinfection of the abdominal skin with iodine and ethanol. B Laparotomy to expose the uterus. C Traditional method: purse-string sutures placed on the uterine wall. D Traditional method: Fetal skin incision created using microsurgical scissors]. E Traditional method: purse-string suture closure. F Modified method: uterine wall punctured directly with a 19G needle under manual stabilization of fetal position. G Modified method: Full-thickness fetal dorsal skin wound created by extending the needle into the fetus; the needle tip is not visible. H Modified method: After 5 s, the needle is withdrawn; due to uterine contraction, minimal or no amniotic fluid is observed. I Maternal abdominal wall closed with skin staples in a single step. J Final disinfection of the incision area with iodine

In Step 2, the traditional method involved purse-string suture placement on the uterine wall (Fig. 1C), uterine incision, fetal skin wound creation using microsurgical scissors (Fig. 1D), and uterine closure via suture tightening (Fig. 1E). In contrast, the modified method omitted sutures entirely. Instead, a 19G needle was used to puncture the uterine wall, amniotic sac (the needle tip was visible in Fig. 1F) and an make 1 mm full-thickness fetal dorsal skin incision (the needle tip was invisible in Fig. 1G) in one step, with minimal manipulation. After needle withdrawal, no obvious amniotic fluid leakage was observed (Fig. 1H). Step 3 (abdominal closure) was completed in both methods using skin staples (Fig. 1I), followed by disinfection (Fig. 1J).

This procedural simplification in the modified protocol significantly improved workflow and reduced surgical complexity.

Hormone-induced superovulation increases fetal number and increase the likelihood of successful pregnancy

To ensure sufficient fetal numbers per pregnancy, we introduced hormone-induced superovulation in the modified protocol (Supplementary Video 2). Hormonal superovulation protocols, using pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG), have been shown to increase the likelihood of successful pregnancy after mating and enhance the number of fetuses per litter, enhancing model efficiency while reducing the number of animals needed. Representative images illustrate less fetal in traditional natural mating (Fig. 2A) versus increased fetal numbers following superovulation (Fig. 2B).

Fig. 2.

Hormone-induced superovulation increases pregnancy efficiency and fetal yield. A Representative photo of a naturally conceived pregnancy with only one fetus in the uterine horn. B Representative photo of a hormonally induced pregnancy showing a significantly larger number of fetuses. C Quantification of total fetal number per pregnancy in natural mating versus hormone-primed mating. D Average number of days between mating and successful detection of pregnancy, with reduced delay in the hormone group. Data are presented as mean ± SEM. * p < 0.05, ** p < 0.01 by Student T test

Quantification across multiple litters confirmed a significant increase in average fetal count per pregnancy using the modified approach (Fig. 2C). Furthermore, the number of mating attempts (defined as plug-positive but non-pregnant events) was markedly reduced (Fig. 2D), highlighting improved mating efficiency.

This hormonal enhancement strategy effectively optimized the reproductive phase of the model, ensuring easier and more efficient pregnancies, increasing fetal number per litter, and reducing breeding variability and animal use.

Modified protocol reduces abortion rate and operative time

We next evaluated maternal and fetal surgical outcomes in pregnancies where both protocols were applied to both uterine horns, fetal survival rates differed significantly between the two approaches. Dead fetuses were frequently observed in the traditional method (Fig. 3A), while viable fetuses with visible India ink–marked wounds were consistently survived from the modified method (Fig. 3B).

Fig. 3.

Modified surgical method reduces abortion rate and operative time. A Representative image of a dead fetus following traditional purse-string surgery. B Representative image of a surviving fetus from the same mother, treated with the modified puncture method; India ink marks the wound site (red box). C Quantification of abortion rate (dead fetuses/total treated fetuses) comparing traditional and modified methods. D Comparison of average operative time per fetus between the two methods, showing a > 80% reduction using the modified protocol. Data are presented as mean ± SEM. *** p < 0.001, *** p < 0.0001 by Student T test

Quantitative analysis confirmed that the modified protocol dramatically reduced abortion rates, with no fetal deaths observed across all litters treated with the modified method (Fig. 3C). Surgical duration was also significantly decreased, from an average of ~ 25 min in the traditional protocol to less than 5 min with the modified approach, representing an > 80% reduction in procedure time (Fig. 3D).

The modified protocol dramatically reduced operative time and significantly lowered the abortion rate, collectively reflecting increased procedural efficiency and enhanced animal welfare, suggesting superior procedural efficiency and improved fetal viability.

Modified protocol maintains consistent wound phenotypes at day 2 post-injury and enables earlier modeling

To determine whether procedural simplification compromised modeling fidelity, we examined wound outcomes at 48 h post-injury. In both traditional and modified protocols, E15.5, E16.5, and E18.5 fetal wounds showed complete re-epithelialization (Fig. 4A–H). These results confirm that the modified protocol reliably reproduces the classic healing phenotypes established by the traditional method.

Fig. 4.

Gross wound healing outcomes at 48 h post-surgery in E15.5 and E18.5 fetuses. A, B Representative dorsal skin wounds at 48 h post-surgery in E16.5 following traditional surgery. C, D Representative dorsal wound appearance and excised tissue from E15.5 fetus at 48 h post-surgery with the modified method. E, F Representative dorsal wounds at E18.5 following traditional method. G, H Representative dorsal wound site and excised tissue from E18.5 fetus at 48 h post-surgery with the modified method. Red box indicates the injury site, which is shown enlarged in the inset

Most studies using the traditional method have only achieved fetal wounding at E16.5, as earlier time points pose greater technical challenges due to rapid fetal development. However, regenerative capacity shifts daily [15], and earlier wounding provides a valuable window for deeper mechanistic insights and therapeutic exploration. Our modified protocol, with its reduced operative time and higher fetal survival, successfully enabled wounding as early as E15.5—one full day earlier than traditional protocols. The gross morphological appearance of E15.5 wounds created using the modified method closely resembled those reported at E16.5 in prior studies, but with an even smoother, more uniform skin surface. This advancement expands the experimental window for investigating early regenerative mechanisms and enhances the utility of the model in skin regeneration research.

These results demonstrate that the modified protocol maintains consistency in reproducing scarless and fibrotic wound phenotypes at early time points.

Scarless and fibrotic phenotypes are maintained at days 7 and 9 post-injury

To further verify that the modified method reliably reproduces the classic phenotypic distinction between scarless healing and fibrotic repair, we extended our observations to day 7 and day 9 post-injury. Figure 5A illustrates the relative size differences among fetal mice at different developmental stages, highlighting the rapid pace of fetal growth.

Fig. 5.

Gross morphology of fetal wound healing at Day 7 and Day 9 post-injury. A Images of E15.5 and E18.5 fetuses subjected to modified surgery and analyzed at extended time points. B, C Representative dorsal wound appearance and excised tissue from E15.5 fetus at (B) 7 days, or (C) 9 days post-surgery with the modified method. D, E Representative dorsal wound appearance and excised tissue from E18.5 fetus at (D) 7 days, or (E) 9 days post-surgery with the modified method. Red box indicates the injury site, which is shown enlarged in the inset

In E15.5 fetuses treated with the modified protocol, wounds were barely visible by day 7 post-injury (Fig. 5B), and by day 9 post-injury, only slight skin depressions remained without any signs of fibrosis or whitening (Fig. 5C). These wounds demonstrated preservation of normal skin texture and color tone, consistent with true dermal regeneration and confirming the absence of scar formation as mentioned above [2].

In contrast, E18.5 wounds treated with the same protocol continued to display prominent white scars at both day 7 (Fig. 5D) and day 9 post-injury (Fig. 5E), with scar areas exceeding the initial wound size, likely due to skin expansion stretching fibrotic tissue during mouse postnatal growth. These wounds also exhibited characteristic features of fibrotic scar formation, including altered skin texture, and loss of normal pigmentation as mentioned above [2].

Together, these findings underscore the robustness of the modified protocol in preserving the age-specific regenerative versus fibrotic healing phenotypes over time. The extended gross morphological tracking further supports that E15.5 skin retains regenerative potential, while E18.5 skin undergoes fibrotic repair—providing a reliable and physiologically relevant model for studying scarless versus scarring wound healing.

Histological evaluation confirms regeneration and fibrosis patterns

Histological analysis at 48 h post-injury further validated the fidelity of the fetal wound model and the effectiveness of the modified protocol. Hematoxylin–eosin staining (H&E) showed mild inflammatory infiltration at the wound sites. In E15.5 skin, the wound revealed re-epithelialization and restored dermal architecture, including regenerated skin appendages such as hair follicles (Fig. 6A, black arrowheads in Fig. 6B). The dermis exhibited normal thickness and collagen fibers organized in a reticular pattern similar to uninjured skin (Fig. 6C, black asterisks in Fig. 6D), with only mild inflammatory infiltration observed at the wound site.

Fig. 6.

Histological analysis of wound sites at 48 h post-injury. A, D Representative H&E (A, B) and Masson trichrome (C, D) staining of E15 fetal skin. Black box indicates the wound site, which is shown enlarged in B, D. Black arrowheads indicate regenerated hair follicles (B). Black asterisk indicates reticular collagen architecture in the dermis (D). Scale bar = 250 μm. E, H Representative H&E (E, F) and Masson trichrome (G, H) staining of E18 fetal skin. Black box indicates the wound site, which is shown enlarged in F, H. White arrowheads indicate revealed fibrotic scarring (F). White asterisk indicates dense, disorganized dermal collagen and absence of hair follicles (H). Scale bar = 250 μm. I Hematoxylin–eosin (H&E) staining confirming correct transverse sectioning. J Quantification of hair follicle regeneration at wound sites. Data are presented as mean ± SEM. ** p < 0.01 by Student T test

In contrast, E18.5 wounds demonstrated features of fibrotic healing: they lacked hair follicles (Fig. 6E, white arrowheads in Fig. 6F), showed disrupted dermal organization, and contained dense, irregular, and hypercellular collagen bundles. Masson’s trichrome staining confirmed prominent collagen deposition in the scar region, highlighted by dense, irregular, dark-blue-stained collagen bundles and loss of normal reticular architecture (Fig. 6G, white asterisk in Fig. 6H). For histological analysis, whole fetuses were sectioned transversely (Fig. 6I).

Quantitative analysis supported these observations: the number of regenerated hair follicles was significantly higher in E15.5 wounds than in E18.5 (Fig. 6J), affirming the regenerative capacity of earlier gestational wounds.

Importantly, based on both gross and microscopic morphology, the modified protocol preserved all four characteristics of regenerative healing at E15.5: (1) absence of dermal fibrosis, (2) retention of skin texture, (3) restoration of appendages, and (4) consistent color tone. In contrast, E18.5 wounds lacked these regenerative features, highlighting the clear phenotypic distinction between scarless and fibrotic healing.

Together, comparison of operative outcomes between protocols reinforced the advantages of the modified method. These results demonstrate that the modified protocol preserves the biological integrity of the fetal skin regeneration model while significantly enhancing maternal and fetal outcomes. It enables precise modeling of age-dependent wound responses, providing a valuable platform for mechanistic studies and therapeutic development in scarless healing.

Discussion

The mouse fetal scarless wound healing model remains a widely used system due to its low breeding cost, short gestational period, and well-established genetic tools for mechanistic studies [14, 17]. Here, we present a modified surgical protocol that improves the traditional fetal wound model [7]. Key updates include: (1) omission of the purse-string suture prior to uterine incision; (2) use of needle tips rather than microsurgical scissors to incise the uterine wall, amniotic sac, and fetal skin; (3) no amniotic fluid supplementation; and (4) closure of the abdominal wall with a skin stapler instead of sutures. Morphological tests verified that our optimized approach reliably produces skin regeneration in E15 (scarless) and scar formation in E18 (fibrotic) fetal wounds, while reducing surgical complexity and lowering abortion rates.

Our optimized approach overcomes technical challenges of traditional models, enhances surgical model success rate, reduce fetal loss, and promotes maternal health. Previous reports have emphasized the technical difficulty, particularly the placement of purse-string sutures in such a small anatomical space [7, 9]. We found that using needle tips enables precise puncture wounds and serves as an effective alternative to microsurgical scissors. Compared to scissor incisions, needle punctures promote self-contraction of the uterine wall and significantly minimize amniotic fluid leakage. This improvement eliminates the need for fluid supplementation, thereby preserving the intrauterine environment and reducing the risk of infection, fetal loss, and miscarriage. Avoiding sutures may further reduce uterine irritation and the likelihood of premature contractions.

Notably, the needle puncture technique simultaneously penetrates the uterine wall, amniotic sac, and fetal skin, preventing fetal displacement often seen in traditional methods after uterine opening. This allows for a smaller surgical field and eliminates the need to externalize most of the uterus to adjust fetal position before surgery. As a result, the risk of uterine tension and irritation during uterine return is significantly reduced, thereby reducing miscarriage and further limiting amniotic fluid loss. Despite these optimizations, careful handling remains essential, and redundant modeling is recommended. From an ethical standpoint, the combined improvements in surgical efficiency, reduced fetal loss, and increased pregnancy predictability align with the 3Rs principle (Replacement, Reduction, and Refinement) in animal research. From an experimental design perspective, lower fetal mortality improves statistical power and reduces variability, enhancing the robustness of downstream molecular or therapeutic studies. In addition, the use of skin staples shortened the surgical time, decreased anesthetic exposure, and minimized abdominal manipulation time.

Importantly, as regenerative medicine seeks to recapitulate fetal-like healing in adult tissues, this optimized model provides a reliable and accessible platform to evaluate candidate pro-regenerative or anti-fibrotic interventions with greater throughput and reproducibility [7, 9, 10, 18–22]. It may also facilitate mechanistic studies of scarring pathways by enabling side-by-side comparisons of interventions in E15.5 versus E18.5 fetuses with consistent baseline phenotypes [23–25].

Surgical access before E15.5 introduces unique technical challenges beyond fetal size. After E12.5, the uterine wall becomes markedly thinner due to decidual regression and reduced myometrial support, rendering it highly prone to tearing under suture tension. This fragility necessitates meticulous control of suture spacing and needle tension to avoid tissue tearing, regardless of gestational age. While uterine wall tension increases with advancing pregnancy, 7–0 nylon sutures were used consistently at both E16.5 and E18.5 in our study, in line with previous reports [12]. Additionally, the fetal membranes at these stages are extremely delicate, increasing the risk of amniotic fluid leakage and membrane rupture during manipulation. The heightened uterine contractility observed before E14.5 further complicates postoperative outcomes, potentially causing placental detachment or embryonic loss. In this context, exo utero approaches are often preferred to avoid uterine closure entirely [26]. However, embryos subjected to exo utero procedures may exhibit developmental abnormalities and delays [27]. Moreover, placental and umbilical vessels are exceptionally fragile and difficult to visualize, making intrauterine positioning and vascular avoidance particularly demanding. These constraints underscore the significant technical refinement required for fetal interventions at early gestational ages and highlight the advantages of our needle-puncture method for creating early-stage fetal skin injuries.

For consistency and direct comparison with the traditional model, which established 1 mm fetal skin wounds with scissors as the standard reference, this study only showed the data obtained by the 19G needle. During optimization of the modified model, we selected needle gauges with outer diameters of ≥ 1 mm to align with the traditional method. Our trials included standard hollow needles (16G–19G) as well as custom-made solid needles ranging from 1–2 mm in diameter. Through these comparisons, we found that the key determinant of procedural success was not needle type alone, but rather precise control of the insertion angle and depth. This highlights the importance of fine motor technique in early-stage fetal surgery and supports the practicality of our simplified approach.

While the modified approach offers numerous advantages, it is not without limitations. First, the precise depth and size of needle-induced wounds may vary slightly between operators. Although this variability appears to have minimal impact on overall wound phenotype in our hands, future refinements such as custom-made microneedle guides could further enhance reproducibility. Second, we limited our evaluation to dorsal skin wounds; whether this protocol is equally effective for limb or facial fetal wounds requires further study.

Conclusion

This study introduces and validates a refined fetal wound healing model in mice that preserves the biological outcomes of the traditional model while substantially improving procedural simplicity, efficiency, and animal welfare. The combination of a simplified surgical technique and hormone-primed mating enables greater scalability and consistency, making it a powerful tool for regenerative biology and fibrosis research. Given the growing interest in translating insights from fetal regeneration into adult therapeutics, this optimized model will be invaluable for accelerating discovery and testing of anti-scarring interventions.

Abbreviations

E16

Embryonic day 16

IACUC

Institutional Animal Care and Use Committee

PBS

Phosphate-buffered saline

H&E

Hematoxylin and eosin

Authors’ contributions

L.H., C.H., and Q.L. conceived experiments; L.H., X.Y., and C.H., conducted the experiments; L.H., and X.Y. wrote the manuscript; L.H. and X.Y. prepared Figs. 1, 2, 3, 4, 5 and 6. All authors reviewed and edited the manuscript. All authors agree to be held accountable for all aspects of this article.

Funding

This project supported by grants from the Science and Technology Commission of Shanghai Municipality Grant No. 22MC1940300 (LQ).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Shanghai Jiao Tong University School of Medicine.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.