Effects of prenatal Doppler ultrasound on the retina of the chick embryo in ovo

Currently, the effect of prenatal ultrasound on foetal development is intensively discussed and the guidelines for prenatal diagnostics have been changed. However, data supporting these concerns are scarce. Therefore, we used an established in ovo model of the chicken embryo to investigate cell proliferation and apoptosis within the retina. A total of 21 chicken eggs were fenestrated on Day 5 and allocated to either the control (n = 8) or exposition group (n = 13). The exposition group was treated with pulsed‐wave Doppler ultrasound (PWD) for 10 min while controls remained without treatment. After subsequent incubation (6–48 h), the eggs were sacrificed, and chicken embryos were examined morphologically (HE‐staining) and immunohistochemically. Counting of apoptotic and proliferating cells per retina was performed using antibodies specific for phospho‐histone‐H3 and active caspase‐3 in combination with a biotin‐labelled secondary antibody and peroxidase conjugated avidin–biotin complex for chromogenic detection. Due to a rather low number of specimens at each time point after ultrasound exposition, we neglected the effects of incubation time and focused on treatment effects. This approach revealed that the median number of proliferating cells is reduced after 10 min of exposure to PWD (569 vs. 766), while the number of apoptotic cells is fairly comparable between groups (5 vs. 6). Our data contribute to a better understanding of prenatal US on foetal development by suggesting that PWD could have an impact on the number of proliferating cells in the developing chicken retina and therefore justify further investigations.

model of the chicken embryo to investigate cell proliferation and apoptosis within the retina.A total of 21 chicken eggs were fenestrated on Day 5 and allocated to either the control (n = 8) or exposition group (n = 13).The exposition group was treated with pulsed-wave Doppler ultrasound (PWD) for 10 min while controls remained without treatment.After subsequent incubation (6-48 h), the eggs were sacrificed, and chicken embryos were examined morphologically (HE-staining) and immunohistochemically.Counting of apoptotic and proliferating cells per retina was performed using antibodies specific for phospho-histone-H3 and active caspase-3 in combination with a biotin-labelled secondary antibody and peroxidase conjugated avidinbiotin complex for chromogenic detection.Due to a rather low number of specimens at each time point after ultrasound exposition, we neglected the effects of incubation time and focused on treatment effects.This approach revealed that the median

| INTRODUC TI ON
Ultrasound (US) plays an important role in prenatal diagnostics for decades and is increasingly used to monitor foetal development and to check for somatic aberrations.Although current maternity guidelines in Germany (Gemeinsamer Bundessausschuss, 2021) recommend one US examinations per trimester, approximately half of the pregnant women underwent at least four US examinations (IQTIG, 2017).
However, there are reasonable concerns regarding the impact of prenatal US on embryonic development but data supporting these concerns are scarce (Linderkamp, 2017).In fact, US impact on cells and tissue and thus might affect foetal development, especially when applied during organogenesis (ter Haar, 2010).On the one hand, US can cause heating secondary to the absorption of acoustic energy in tissues.On the other hand, indirect non-thermal mechanisms occur with cavitational and non-cavitational effects, which may lead to an increased temperature but also damaged tissue structures (Abbott, 1999;Nowicki, 2020).Therefore, the thermal index (TI) and mechanical index (MI) have been established as guideline values for US diagnosis that should not be exceeded during the examination (Sande et al., 2021).These indices are relative indicators, for example of a possible rise in temperature along the propagation axis of the ultrasonic beam.The soft-tissue TI (TIS) provides information about the temperature increase inside soft, homogeneous tissues.Pellicer et al. (2011) demonstrated that pulsed-wave Doppler US (PWD), a widely used US technique with a high acoustic output, results in cellular damage of foetal liver in rats.Hence, for example in eye scanning applications, it is recommended that TIS not exceed a maximum value of 1.0 (Herman & Harris, 2002;Nowicki, 2020).
Whether harm to mother and child occurs due to an US examination depends on several factors: The shorter the exposure duration and the lower the applied acoustic energy or power, the lower the risk of adverse effects (Kollmann et al., 2013).For this reason, according to the recommendation of the European Committee of Medical US Safety, the focus should not be just only on TI and MI, but also on a proper indication according to the ALARA principle (ALARA = As low as reasonably achievable; European Committee of Medical Ultrasound Safety (ECMUS), 2008).Due to the lack of studies regarding the impact of US on foetal development in the first trimester, no proper risk assessment is currently possible.Therefore, we focused on the impact of PWD on the early retinal development of the chicken embryo.This is an established animal model, especially for in ovo studies of embryonic eye development and the morphology of the retina (Klose et al., 2017;Lindner et al., 2017;Schneider-Kolsky et al., 2009;Streckenbach et al., 2019Streckenbach et al., , 2022)).
At developmental Day 5 (D5), fertilized eggs were checked with an egg candler (Albert Kerbl GmbH) regarding vascular development and movement of the embryo.At this stage, there were 24 fertilized eggs with normal development.Subsequently, eggshells were fenestrated under sterile conditions and closed with Tegaderm™ plast (3M™ Germany GmbH).Viability of the embryos was examined 24 h later by using egg candler.Vital embryos were randomly allocated to the control or treatment group, that is the application of PWD for 10 min.After US exposition, eggs were incubated for an additional period of 6, 12, 24 and 48 h (Figure 1).Embryonic development was then terminated by cooling the eggs for 60 min in crushed ice.Embryos were collected, and the head and body were preserved separately in 4% buffered formaldehyde and embedded in paraffin for histological examination.
The experiments were compliant with national animal welfare legislation.

| Ultrasound exposition
A colour-coded state-of-the-art US system (Aplio i900, Canon, Canon Medical Systems) using a 11-MHz linear array transducer (11MC4/ PVT-712BT) was employed to perform two-dimensional B-Mode and PWD scans in a standardized manner, that is with the transducer mounted to a clamp system (Thomann GmbH; Figure 2a).Eggs were transferred from the incubator to an egg holder and scanned once for 10 min at 23°C of room temperature.Pre-warmed ultrasonic gel was used to allow oblique probe adjustment to image the number of proliferating cells is reduced after 10 min of exposure to PWD (569 vs. 766), while the number of apoptotic cells is fairly comparable between groups (5 vs. 6).Our data contribute to a better understanding of prenatal US on foetal development by suggesting that PWD could have an impact on the number of proliferating cells in the developing chicken retina and therefore justify further investigations.embryonic eyes (Figure 2b).Depending of the different location of the embryos inside the egg, the scanning depth varied between 1.0 and 2.5 cm with activated autofocus.According to the data provided by the manufacturer, the maximum attenuated spatial peak average intensity (I SPTA .3 ) was 85 mW/cm 2 for B-mode and 545 mW/cm 2 for PWD-mode (Canon Medical Systems, 2020).These values are well within the recommended power outputs for foetal and neonatal use, with a maximal recommended level of 720 mW/cm 2 (Ng, 2002).Mechanical index and TI were displayed on the US screen and evaluated in real-time during both B-and PWD-exposition.
The B-and PWD-mode were used to visualize the embryonic heart and ensure the viability of the embryo and the correct direction of the US beam.The embryos in the control group were exposed to the same procedures, but with the probe being switched off.

| Histological and immunohistochemical evaluation
Serial sections from the heads (4 μm) were mounted on Superfrost®slides (Thermo Fisher Scientific, Inc.), and every slide was submitted to staining with haematoxylin/eosin (HE) according to standard procedures (Figure 3).Slides were screened at 100× magnification (Leica DMI 4000; Leica) for the presence of the eye and the optic nerve, and adjacent sections were used for immunohistochemical detection of proliferating and apoptotic cells, respectively.
Briefly, sections were deparaffinized in xylene and hydrated through a series of graded alcohols before blocking of endogenous peroxidases (POD).Antigen retrieval was achieved by treatment with 10 mM citrate buffer (pH 6.0) using a microwave (3 × 7 min, 450 W).
Unspecific binding sites and endogenous biotin were blocked with normal horse serum (Vectastain Elite ABC Universal Plus Kit Peroxidase; Vector Laboratories, Inc.) and avidin (Avidin/Biotin Blocking Kit; Vector Laboratories, Inc.).For quantification of proliferating and apoptotic cells, antibodies specific for phospho-histone-H3 (Ser10) (Merck KGaA) and active caspase-3 (R&D Systems) were used after being diluted in horse serum as appropriate.For detection of bound antibodies, a biotin-labelled secondary antibody in combination with peroxidase-labelled avidin-biotin complex (Vectastain Elite ABC Universal Plus Kit Peroxidase; Vector Laboratories, Inc.) was used.
Finally, slides were incubated with 3,3′-diaminobenzidine tetrahydrochloride (DAB) for visualization of phospho-histone-H3 and caspase-3-positive cells, respectively.Counterstaining was omitted, and the number of stained cells and retinal area was recorded for each retina.
In the case of phospho-histone-H3-positive cells, the QWin software package (Leica Microsystem) with predefined thresholds was employed, while caspase-3-positive cells were counted manually.
In both cases, the examiner was blinded to the origin of the tissue samples.
F I G U R E 1 Timeline of the experiment.Twenty-one eggs were divided into two groups: Eight eggs were just fenestrated (control group) and 13 eggs were fenestrated and exposed to PWD for 10 min (exposition group).Chicken eggs from the two groups were incubated for a period of 6, 12, 24 and 48 h, and then, incubation was terminated for histological workup.Of the 24 eggs designated for the study, three were excluded because development stopped after fenestration.The remaining 21 eggs were allocated to the control (n = 8) and treatment (n = 13) groups.Despite the concomitant downsizing of the predefined treatment groups, we decided to keep the incubation time as planned, that is incubation for 6, 12, 24 and 48 h after US exposition.Incubation and PWD examination of all remaining 21 eggs were successful.
Mechanical index and TI were displayed on the US screen and evaluated in real-time during both B-and PWD-mode exposition with a variation between 0.9 and 1.2.Immunohistochemical and histopathological evaluation was successfully performed on all 21 paraffin-embedded embryos, namely 13 chicken embryos from the exposition group and eight embryos from the control group (Figures 3-5).An example for phospho-histone-H3 and caspase-3-positive cells is presented in Figures 4 and 5, and the results of the quantitative analysis are summarized in Table 1.
There was a range of phospho-histone-H3 cells from 0 to 757 cells/mm 2 in the exposition group and from 263 to 687 cells/mm 2 in the control group.By using Mann-Whitney U test, the exposition group showed no significantly lower number of phospho-histone-H3-positive cells compared to the control group (Table 1, Figure 6).
The number of caspase-3-positive cells ranged from 2 to 14 cells/mm 2 in the exposition group and from 2 to 10 cells/mm 2 in the control group.There were no significant differences between the exposition and control groups regarding the number of apoptotic cells (Table 1, Figure 6).Due to a rather low number of specimens at each time point after US exposition, we neglected the effects of incubation time and focused on the exposure effects.This approach revealed that the median number of proliferating cells is reduced after 10 min of exposure to PWD (398 [min. 0;max. 757] vs. 593 [min. 263;max. 687]), while the number of apoptotic cells is fairly comparable between exposition group 5 (min.2; max.14) and control group 4 (min.2; max.10).This indicates that there may be a lower number of proliferating cells in the retina of embryos exposed to US.

| DISCUSS ION
Ultrasound has played an important role in prenatal diagnostics for decades and its importance continues to increase due to technical developments, which are leading to appropriate therapeutic strategies or to predict foetal outcome (Micu et al., 2018).Currently, the effect of prenatal US on foetal development is intensively discussed and the guidelines for prenatal diagnostics have been changed.Data supporting these concerns are scarce.In an extensive evaluation regarding safety of prenatal US diagnostics for the child, it was demonstrated that in just three out of 39 epidemiologic studies analysed, prenatal US diagnostics were performed in the first trimester using high-energy US techniques such as PWD.However, especially in the first trimester, for example brain development and migrating neurons are highly susceptible to any disruptive influences like mechanical vibrations caused by US (Linderkamp, 2017).
The results of our present study indicates that PWD may have an effect on the number of proliferating cells in the developing chicken retina.Although we did not demonstrate that PWD significantly reduced the number of proliferating cells, there were 593 (min.263; max.687) proliferating cells/mm 2 in the control group versus 398 (min.0; max.757) proliferating cells/mm 2 in the verum group (Table 1, Figure 6).There was no difference in the number of apoptotic cells between both groups (verum: 5 [min.2; max.14]; control: 4 [min.2; max.10]).Chicken eggs with a longer duration of incubation (24-48 h) showed a higher number of proliferating cells than those which were sacrificed after 6 or 12 h.Because of the low number of viable chicken eggs for each timepoint after US exposition, the testing for statistical significance is not expedient.Furthermore, this observation is to be expected since the retinal cells proliferate and divide with increasing duration of the incubation.
To conclude from the experiments concerning the influence of the US on the development of the retina despite a low number of cases, we had to neglect the effect of the incubation time (6, 8, 12, 24 or 48 h).Manipulation could initiate the apoptotic process within the chicken eggs as shown by the loss of eggs during the experiment.It is possible that the apoptotic process also started within the viable chicken eggs throughout the incubation time with or without the use of US.In addition, eggs had to be tilted for US exposition, which increased the risk of infection because the albumen came into contact with the Tegaderm™ plast.This could be an explanation for the high number of caspase-3-positive cells/mm 2 in both verum and control group.
K E Y W O R D S chicken embryo, prenatal ultrasound, retinal development | 1005 KÖNIG et al.

F
Ultrasound exposition.(a) Positioning and attachment of the transducer on the fenestrated egg.(b) PWD of the chicken embryo displaying arterial blood flow in the heart.

F
Chicken embryo head (HE stain) with lens (1), optic nerve (2) and retina (3).F I G U R E 4 Immunohistochemical evaluation of the chicken embryo retina (a).A specific antibody for active caspase-3 was used to determine the number of apoptotic cells (black arrows in enlarged view) (b).| 1007 KÖNIG et al.

F
I G U R E 5 Immunohistochemical evaluation of proliferation cells in the chicken embryo retina (a).For quantification of proliferating cells, we used a specific antibody for phospho-histone-H3.Phospho-histone-H3-positive cells in a retina (black arrows in enlarged view) were counted using Leica QX-Count (b).