Low frequency vibrations disrupt left-right patterning in the Xenopus embryo

We propose that low frequency vibrations are an alternative to pharmacological treatments that target the cytoskeleton, because unlike exposures to drugs and chemicals, we know exactly when in development the embryo is subjected to treatment. Vibration therefore allows discrete developmental periods to be targeted, and also allows multiple periods to be tested in the same embryos (Figure 3). Additionally, the effects of low frequencies are subtle enough that vibration can be paired with other treatments (Figure 6). In our experience, multiple treatments, each at sub-toxic levels, are typically lethal when combined (Vandenberg & Levin, unpublished data); thus, we have now achieved an experimental system where the effects of altered cytoskeletal dynamics can be paired with treatments targeting other pathways. Of course, vibration protocols are not expected to completely replace pharmacological treatments, but do provide an alternative to these methods when there are concerns about the timing of chemical exposures or when multiple treatments are necessary.

Importantly, knowing exactly when the vibration treatments have ended is not informative about the period of time it takes the embryo to recover from treatment. In cultured cells, the cytoskeleton remains disorganized several hours after vibration, but recovers within 24-hours; cell proliferation resumes sometime after that point [54]. Thus, it is important to address whether the effects of vibration are due to early interruptions of the LR pathway, or whether they could be indicative of effects on other cytoskeletal structures (like cilia) later in development. There are two results that address this issue. First, the striking differences between embryos vibrated from 1 cell through neurulation, compared to embryos vibrated from 2 cell through neurulation, clearly indicates that an extremely early event is affected by this treatment (Figure 2); no differences in the effects on cilia would be expected from these treatments. Second, several treatments that start prior to and span the period of time where ciliary motion and nodal flow occur do not significantly affect the LR axis (i.e. vibration from st. 8 through neurulation; Figure 2). If vibration were affecting the organization or assembly of cilia, these treatments, which fully encompass the developmental periods where cilia appear and flow begins [86], [87], [88], should be effective. Instead, all of our results are consistent with vibration affecting early LR pathways via two mechanisms: disruption of the chiral cytoskeleton and altered epithelial integrity.

In summary, we have identified a new treatment paradigm by which the LR axis of the Xenopus embryo can be disturbed, and our results suggest that the physical perturbations of low frequency vibrations affect the very early cytoskeleton (during the first cleavage stage) and tight junctions. Thus, vibration acts to disrupt two specific steps in the early LR pathway, and ultimately alters patterning of this axis. For the first time, we have identified a treatment that can separately randomize the alignment of the LR axis with the other two axes in addition to disrupting the amplification and restriction of LR information, thus producing different LR phenotypes depending on when the exposures occurred. This information should put scientists in this field one step closer to understanding how the LR axis is consistently oriented and organ concordance is controlled. The question of how symmetry is broken remains a fascinating and provocative question with important relevance to the field of developmental biology as well as medicine, and biophysical tools including the low frequency vibrations explored here offer tremendous power to explore the many puzzles that remain unsolved.

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by Tufts University's Institutional Animal Care and Use Committee (#M2008-08). Xenopus laevis embryos were collected and fertilized in vitro according to standard protocols [89] in 0.1× Modified Marc's Ringers (MMR) pH 7.8+0.1% Gentamycin. Xenopus embryos were housed at 18–23°C and staged according to [90].

4-inch Sony speakers (Model # 1-544-670-11) were connected to a Gwinstek GFG-8216A function generator. Placed on these speakers were embryos (100–200 per treatment) in polystyrene petri dishes (Fisherbrand Catalog # 0875712, diameter = 10 cm, depth = 1.5 cm) containing approximately 40 ml 0.1× MMR. These dishes were vibrated at specific frequencies over various developmental time periods (as indicated in the text). Typically, two speaker systems were positioned on a benchtop near each other, and both were operated at the same time and at the same frequency, set to vibrate at the highest non-lethal amplitude on the sine wave setting for all experiments. When the function generator was set at 7 Hz, the average acceleration of the dish was measured as 0.53 m/sec2 using a calibrated Endevco accelerometer (Model 2250K). At 15 Hz, the average acceleration of the dish was 3.96 m/sec2. The ambient temperature was always ≤23°C, a temperature that does not affect orientation of the LR axis. Constructive or destructive interference did not appear to occur between the two speakers, and similar results were obtained with single speakers. In each experiment, non-vibrated control dishes were placed in an incubator away from all external sources of vibration.

Embryos were placed on 4-inch Sony speakers after removal of the stationary magnet, leaving a speaker that produced electromagnetic fields (EMFs) in the absence of physical vibrations. The speakers were connected to a frequency generator, and all other aspects of the experiment were the same as in the vibration studies.

At stage 45, live Xenopus embryos were analyzed for position (situs) of the heart, stomach and gall bladder according to [76]. Heterotaxia was defined as the reversal in position of one or more organs including situs inversus (reversal of all three organs), which was also considered separately in some experiments. Only embryos with a normal dorsoanterior index (DAI = 5) were scored to prevent confounding of randomization caused by midline defects [91]. Percent heterotaxia was calculated as number of heterotaxic embryos divided by the total number of scorable embryos. A χ2 test was used to compare absolute counts of heterotaxic embryos. 10% was set a priori as the minimum meaningful difference due to the rate of spontaneous heterotaxia observed in untreated controls (1–3%) and the rates of heterotaxia observed in well-characterized mutants and treatments [17]. For all treatments, experiments were performed in duplicate or triplicate, and the data presented are the sums from those replicates.

Whole mount in situ hybridization was performed using standard protocols [92]. In situ hybridization probes against Xnr-1 (the Xenopus nodal) mRNAs [64] were generated in vitro from linearized template using DIG labeling mix (Roche). Embryos vibrated for multiple stages were examined for Xnr-1 localization. A χ2 test was used to compare absolute counts of embryos with correct versus incorrect Xnr-1 expression.

All embryos with heterotaxia were separated into one of three groups: single organ inversions, double organ inversions, and situs inversus. The percentage of each group was calculated for several vibration periods as well as previously characterized reagents that affect LR orientation (Supplemental Table 2). Each treatment was plotted on a ternary graph using these three measures according to methods described previously [67]. Using the calculations provided by the ternary plot program, https://webscript.princeton.edu/~rburdine/stat/three_categories, when there was no overlap of confidence intervals, results were considered statistically significant. When confidence intervals overlapped, a χ2 test was used to calculate statistical significance.

For these experiments, embryos were vibrated at 7 Hz, treated with drugs targeting specific LR relevant pathways, or both starting at the 1 cell stage. For drugs concentrations, the following doses were utilized: 2.1 µM lindane [75]; 13 nM nocodazole [34]. All treated embryos remained in the drug solution overnight until stage 12 of development when they were repeatedly washed and then housed in clean 0.1× MMR for the remainder of the experiment. At st. 45, embryos were scored for heterotaxia and the results from each treatment were compared to determine whether there was an additive effect of vibration and drug.

A tight junction assay was performed according to standard protocols [40]. Briefly, control and vibrated embryos were collected at st. 6, cooled to 10°C for 10 min, and then incubated in a 1 mg/ml solution of EZ-Link Sulfo-NHS-LC biotin in 0.1×MMR and 10 mM HEPES (pH = 7.8) at 10°C for 10 min. This biotin molecule has been modified to prevent it from passing through normal tight junctions unless epithelial integrity has been compromised. After labeling, embryos were rinsed twice with cool 0.1×MMR, fixed in MEMFA, processed and embedded in paraffin. 5 µm sections were cut on a Leica M2255 microtome and mounted on glass slides.

The slides were deparaffinized, rehydrated, and blocked with 10% goat serum in 1×PBS for 1 hr at room temperature, then incubated with the fluorescent secondary antibody streptavidin-Alexa 555 at a concentration of 1∶200 in blocking solution. Slides were then washed, mounted with a coverslip, and imaged.

Immediately following fertilization, 1 cell embryos were placed in 3% Ficoll and injected with a 1∶1∶2 mixture of Oregon Green (OrG, MW = 10 kDa), Tetramethylrhodamine (TMR, MW = 70 kDa), and RODI water using standard protocols (100 ms pulse time and borosilicate glass needles with a bubble pressure of 60–70 kPa). Injections were uniformly delivered into the middle of the embryo at the animal pole of the embryo. Injected embryos were then either vibrated at 7 Hz for various intervals of time within the first 2 cell divisions or stored in a location with equivalent temperature but no source of external vibration. Following treatment, embryos were fixed in MEMFA, washed, and embedded in agarose according to standard protocols [93]. 100 µm sections were cut using a Leica VT1000S vibratome and imaged in PBS.

An Olympus BX-61 with a Hamamatsu ORCA AG CCD camera, controlled by MetaMorph software, was used for all fluorescence imaging. For all other experiments, images were collected with a Nikon SMZ1500 dissection microscope with a Retiga 2000R camera and ImageQ software. Photoshop™ was used to orient, scale, and improve clarity of images. Data were neither added nor subtracted; original images are available on request.

The authors gratefully acknowledge Punita Koustubhan, Amber Currier, and Claire Stevenson for Xenopus husbandry and general laboratory assistance, Dr. Robert D. White for acceleration measurements, Dr. Joan Lemire for comments on the manuscript, and members of the Levin Lab for helpful discussions. We also thank Chris Wright for the Xnr-1 probe and are grateful to Gary Leisk, Nereus K. Patel and William G. Temple for construction of vibration device prototypes.