Scanning Various Nanomaterials and Variable Camouages to Evaluate the Biocompatibility in Multi-organs of Embryos

Background: Nanomaterials are under a wide range of application prospects in human health and disease, such as medical imaging and drug delivery and treatment. With the gradual advent of the nano era, the safety of nanomaterials in biology must be evaluated more widely and deeply. Therefore, supercial safety assessment based on specic cell lines in vitro is dicult to meet the current needs. Multi-organs assessment in vivo will be benecial in establishing a more comprehensive pathway for the understanding of nanomaterials-induced biotoxicity. In this work, we employed a series of genetically modied zebrash models for nanomaterials toxicity assessment. The results demonstrated that cadmium selenide (CdSe) quantum dot (QDs) was the most toxic after scanning some popular nanomaterials. Among modied methods of silica coating, core-shell structure development and organic molecular camouage, the polyethylene glycol camouage method exhibits a better performance of improving the biocompatibility of CdSe QDs in multiple-organs including hearts, nerves, blood vessels, and immune system. Our work provides an alternative paradigm for more in-depth and sensitive preclinical validation of biocompatibility, especially in neurotoxicity and cardiotoxicity of embryos, and a guidance for reducing the toxicity of biomedical materials.


Introduction
Due to the small size effect, surface effect, quantum size effect and macro quantum tunneling effect of nanomaterials, they present optical, acoustic, electrical, magnetic and temperature sensitive features that conventional materials do not. [1,2] With the rapid development of many subjects, nanomaterials with excellent characteristics have a wide range of applications in disease treatment, medical imaging, biological detection, biosensing, etc. [3] Such as nanoenzymes in the elimination of ROS, [4] and gold nanoparticles (AuNPs) in biological detection, [5] the application of up-conversion nanoparticles (UCNPs) in imaging and tumor diagnosis, [6] quantum dots (QDs) as uorescent markers widely used to image and track the migration of drugs in the body. [7,8] The inevitable contact between humans and nanomaterials has made the safety of nanomaterials attract much attention all over the world. [9] The small size of nanoparticles makes it easier to enter cells or organelles through the gaps in the biological membrane, and combine with biological molecules or catalyze chemical reactions. The complex interaction could change the structure and the function of biological macromolecules and physique and ultimately lead to cellular cell and tissue toxicity. Some nanoparticles may trigger interactions with blood, affect major organs or accumulate in organs, and even cause immune responses. [10][11][12] In many current studies, the toxicity veri cation of nanomaterials is mainly based on the physiological activity test of speci c cell lines in vitro, such as proliferation and apoptosis. [13,14] Part of the evaluation involves body weight changes and macroscopic pathological changes of main organs, such as direct naked eye observation of main organs. [15,16] Even when it comes to the pathological changes presented by Hematoxylin-eosin staining at the cellular level, it can only be appeared after a large amount of toxic substances are accumulated in the body. The sensitivity of toxicity detection and the comprehensive multiple organ toxicity evaluation are severely restricted by the above means, especially in neurotoxicity and cardiotoxicity. Therefore, more sensitive and systematic evaluation methods should be urgently employed to assess the biocompatibility of biomaterials in vivo. Zebra sh has been recognized as important model for human disease which has about 70% similarity with humans. [17] Since zebra sh has regeneration capability in its organs such as ns, central nervous system (CNS), heart, pancreas, liver, and kidney, it has been used for different models of injury for example in cardiovascular, neurological, and metabolic diseases. [18][19][20] The technology of transgenic and genome editing is mature and e cient in zebra sh, which made zebra sh as a common model for genetic screen and disease models. [21][22][23] Moreover, the zebra sh embryos are transparent and develop externally, a variety of imaging modalities were applicable to live image of embryogenesis and disease occurrence. [24][25][26] With the help of uorescence microscope and transgenic uorescent reporters, researchers could speci cally mark a variety of tissues such as vessels, neurons, in ammatory cells and examine the structural and functional phenotypes associated with genetic mutations and chemical exposure in vivo. [27] Hence, zebra sh embryos has been extensively applied for safety evaluation of medicines in preclinical studies because of its sensitivity to harmful substances and easy genetic modi cation. [28][29][30] In this study, a variety of zebra sh embryos have been employed to scan the safety of popular nanomaterials. Among the widely used biomedical nanoparticles (AuNPs, UCNPs, C QDs, ZnO QDs and CdSe QDs), CdSe QDs indicated the most severe physiological toxicity, which caused cardiac dysfunction, vascular and neurons damage and triggered in ammation in zebra sh embryos. In order to reduce its toxicity and make it safer to use in organisms, CdSe QDs was camou aged by wrapping with silicon dioxide (SiO 2 ), zinc sul de (ZnS), polyethylene glycol (PEG). It is worth noting that PEG-wrapped CdSe QDs could signi cantly increase the survival rate, reduce the deformity rate. After the comprehensive evaluation of a series of transgenic zebra sh models obtained earlier. [31,32] We were surprised to nd that the toxicity of CdSe QDs to nerve, heart and blood vessel could be signi cantly reduced or even eliminated after PEG modi cation. Our approach might inspire future evaluation strategies of biocompatibility for biomedical nanoparticles before basic studies and clinical applications, especially in nerves, hearts and blood vessels of embryos that are particularly sensitive to harmful substances.

Characterization and screening toxicity nanomaterials
Gold nanoparticles, up-conversion nanoparticles, and quantum dots have a wide range of applications in imaging technology and treatment of disease, but it will have to be monitored more rigorously for toxic and side effects before they can actually be used for biomedical purposes. We prepared AuNPs, UCNPs, C QDs, ZnO QDs, and CdSe QDs to assess their toxic effects on zebra sh embryos. Through the analysis of dynamic light scattering (DLS), the size of AuNPs, UCNPs, C QDs, ZnO QDs and CdSe QDs are approximately 13 nm (Fig. 1A), 40 nm (Fig. 1B), 6 nm ( Fig. 1C), 8 nm (Fig. 1D), 14 nm (Fig. 1E). The excitation wavelength of C QDs is 405 nm and the emission wavelength is 570 nm ( Figure S1). The excitation wavelength of ZnO QDs is 405 nm, and the emission wavelength is 580 nm ( Figure S2). The excitation wavelength of CdSe QDs is 405 nm, and the emission wavelength is 600 nm ( Figure S3). To explore their respective safety, the 4 hours post fertilization (4hpf) zebra sh embryos were incubated in hatching solution E3 with 0.2 mg/mL, 0.15 mg/mL, 0.1 mg/mL, 0.05 mg/mL nanomaterials respectively and analyzed by survival rate and hatching rate. The results showed all zebra sh in the 0.2 mg/mL and 0.15 mg/mL ZnO QDs and CdSe QDs died, while the survival rates of the AuNPs, UCNPs, and C QDs group were about 80%-90% ( Fig. 1F-I), which suggested that AuNPs, UCNPs, and C QDs are less toxic than ZnO QDs and CdSe QDs (Fig. 1F). When the zebra sh embryos were incubated in 0.1 mg/mL, the survival rate and incubation rate of CdSe QDs was 60% which was the lowest among these nanomaterials ( Fig. 1J, K). While the concentration was reduced to 0.05 mg/mL, there was no signi cant difference in zebra sh survival rate and hatching rate after nanomaterials incubation (Fig. 1L, M). Taken together, our results demonstrated that CdSe QDs is the most toxic among these nanomaterials to zebra sh embryos and higher than 0.1 mg/mL will affect the zebra sh development including survival and hatching. Although some nanomaterials are highly toxic, how to improve their biocompatibility has become an important concern due to their unique properties and need to be used in biomedicine.

Toxicity of CdSe QDs after different camou age methods
In order to decrease the toxicity of CdSe QDs and improve their application in biomedicine, we have selected commonly used modi cation methods and tested their toxicity reduction effects. We chose three modi cation methods and the principal was to wrap CdSe QDs with SiO 2 , ZnS and PEG. Because SiO 2 has better biocompatibility, we believed that SiO 2 wrapped CdSe QDs can decrease the toxicity of CdSe QDs ( Fig. 2A). The second was a widespread modi cation method of CdSe QDs, which is to wrap ZnS around it to prepare CdSe@ZnS core-shell structure. According to previous literature reports, this method could reduce the toxicity of CdSe QDs (Fig. 2B). In the third type, almost non-toxic PEG was utilized to modify CdSe QDs (Fig. 2C). Since PEG modi cation was widely used in the modi cation of a variety of biomaterials to improve its biocompatibility and reduce toxicity. The particle size analysis and transmission electron microscopy (TEM) images showed that the success of these three modi cation methods ( Fig. 2A-F). To evaluate different modi cation methods on zebra sh embryos toxicity, we rst tested the different modi cations on zebra sh survival rate, hatching rate and malformation rate (body bending) ( Figure S4). Built on the previous results, we choose the concentration of 0.1 mg/mL for the next study. The zebra sh embryos were incubated in E3 solution with CdSe QDs, CdSe@SiO 2 , CdSe@ZnS and CdSe@PEG and DMSO as control. Results indicated that CdSe@PEG group gets the highest survival rate was 100%, hatching rate was 100% and low malformation rate was 0 compared with CdSe@SiO 2 were 80%, 70%, 10%, CdSe@ZnS group were 75%, 70%, 50%. CdSe@SiO 2 , CdSe@ZnS group has similar survival rate, while CdSe@ZnS group has lower hatching rate was 70% and higher malformation rate was 50% ( Fig. 2G-I). These different modi cation methods analyses display that CdSe QDs with PEG modi cation signi cantly reduced its toxicity during zebra sh development. These results also suggest that PEG modi cation is a better way to improve the biocompatibility of nanoparticles, although the camou age of silicon dioxide and core-shell structure encapsulation can reduce the toxicity of nanoparticles.

Cardiotoxicity of CdSe QDs before and after PEG camou age
The heart is the main organ supplying blood to the whole body, so nanoparticles entering the body will be the rst to enter the heart through the blood circulation. The embryonic heart is also the most active and sensitive period of life, so cardiotoxicity should be tested rst. CdSe QDs has a certain effect on zebra sh cardiovascular development. We rst detect the CdSe QDs and PEG-modi ed CdSe QDs on heart morphological development. Results displayed that CdSe QDs incubation caused obvious pericardial edema was 93% at 3 days post fertilization (3dpf) compared with the control group (Fig. 3A, B). While PEG-modi ed CdSe QDs signi cantly reduce the pericardial edema caused by CdSe QDs from 93-10% (Fig. 3A, B). In addition, we found the CdSe QDs before and after modi cation did not have a serious effect on the heart rate ( Supplementary Videos 1, 2, 3 and S6). Therefore, our results demonstrated that the CdSe QDs coated with PEG is an effective way to rescue the pericardial edema caused by CdSe QDs.

Vascular toxicity before and after PEG camou age for CdSe QDs
Blood vessels are the transport channels of nanoparticles into the body. It has been reported that zebra sh embryos exposure to CdSe from 6 to 96 h exhibited vascular malformations. Also, CdSe QDs was highly toxic on the heart and when nanomaterials were injected intravenously, they quickly entered into blood vessel. Therefore, it is necessary to explore whether PEG-modi ed CdSe QDs can decreased the vascular toxicity of CdSe QDs. To visualize the vascular effects, the transgenic line Tg (Kdrl: mcherry) zebra sh with endothelial cells speci c Kdrl promoter driving mcherry uorescent was adopted. When vascular endothelial cells were affected, its expression decreased and showed vascular malformations. The whole images of zebra sh vascular endothelial cells showed that CdSe QDs incubation signi cantly reduce the expression intensity of vascular endothelial cells in both head and trunk (Fig. 4A-D). It is worth noting that CdSe QDs also lead to partial vessels missing (Fig. 4A, B; arrow indication). Yet, notable improvements in vascular marker expression and vascular integrity were observed for PEG-modi ed CdSe QDs in comparison with CdSe QDs (Fig. 4A-D). Vessel density analysis results indicated that the cardinal vein density signi cantly reduced by more than 50% in the CdSe QDs group compared the control group (Fig. 4E, F). It is worth noting that increased in the PEG-modi ed CdSe QDs increased about 32% compared with CdSe QDs group (Fig. 4E, F). Besides, the intersegment vessel of zebra sh tail indicated abnormal bending or missing in the CdSe QDs group, which could be partly rescued in the PEG-modi ed CdSe QDs group ( Figure S5). All these results suggested that PEG-modi ed CdSe QDs was an effective modi cation to reduce the in uence of CdSe QDs on angiogenesis.
2.5. Effect of CdSe QDs before and after PEG camou age on in ammatory response A variety of materials will be treated as foreign bodies by the immune system after entering the body, the most typical is to activate the in ammatory response. We veri ed whether CdSe QDs caused in ammation in zebra sh embryos with the transgenic line Tg (coronin1a: EGFP) zebra sh embryos which the macrophages and neutrophils was marked by green uorescent protein (GFP). Upon injury or infection to living tissues, in ammatory cells attracted by chemokines and cytokines are released from dead or dying cells and recruited to the injury site. As shown in Fig. 5, the expression intensity of macrophages and neutrophils in the CdSe QDs treatment group increased nearly twice (Fig. 5A-D). In addition, the number of macrophages and neutrophils increased sharply with CdSe QDs incubation ( Fig. 5A-C). Compared with the CdSe QDs, the CdSe@PEG caused an obvious reduction in the macrophages and neutrophils intensity as well as number (Fig. 5A-D). These data demonstrated that PEG-modi ed CdSe QDs could alleviate the macrophages and neutrophil-mediated systemic in ammation in Tg (coronin1a: EGFP) zebra sh triggered by CdSe QDs. This provides a new proof for reducing the immunotoxicity of nanomaterials.

Neurotoxicity of CdSe QDs before and after PEG camou age
This is a common phenomenon that there is little mention of any neurotoxicity related safety veri cation in the massive papers in the eld of materials. More recently, with the occurrence and development mechanism of neurological diseases gradually discovered, the potential nervous system damage caused by nanoparticles must be prudently monitored. To investigate whether CdSe QDs have the toxicity on neurons, we used the Tg (elavl3: Gal4vp16, UAS:mCherry) to detect the neurons expression, in which the pan-neuronal HuC/Elav3 promoter drives mCherry expression. When neurons damaged, the embryos exhibited reduced uorescence expression and abnormal behavior. Results showed the neurons expression signi cantly reduced in the zebra sh head and spinal cord especially in midbrain and hindbrain in the CdSe QDs group compared with the control (Fig. 6A-D). While the midbrain and hindbrain neurons expression indicated the opposite trend in the CdSe@PEG group (Fig. 6A-D). In addition, the spinal cord neurons expression was obviously increased in CdSe@PEG group but not reach the control level (Fig. 6A-D).
To further examine the in uence of CdSe QDs and CdSe@PEG on the nervous system, we examined the behavior of zebra sh after different treatments. During the 30 min recording period, the trajectories of zebra sh in the CdSe QDs group were shorter and they tend to be more stationary while the trajectories were longer in the control and CdSe@PEG (Fig. 7A, C). In the heat map of the movement trajectory, it was more intuitive to nd that the zebra sh in the CdSe QDs group prefer to be still, and the CdSe@PEG group reversed this phenomenon and the zebra sh became active (Fig. 7B). The statistics of moving speed (Fig. 7D),active time (Fig. 7E) and active frequency ( Figure S7) and resting time ( Figure S8) indicated that CdSe QDs signi cantly reduces the movement speed, active time and active frequency of zebra sh, which can be rescue by PEG-modi ed CdSe QDs. Taken together, PEG-modi ed ZnO QDs CdSe QDs could reduce CdSe QDs induced neurotoxicity and abnormal behavior.

Conclusions
In summary, we selected a series of nanomaterials, such as gold nanoparticles, upconversion nanomaterials and quantum dots, which are most commonly used in the diagnosis and treatment of diseases in vivo in recent years. The zebra sh embryos, which is a very sensitive animal model to toxins, is employed to scan their safety based on zebra sh mortality and hatching rate. Our results reveal that the toxicity of CdSe QDs is higher than that of gold nanoparticles and upconversion nanomaterials. In vivo organ toxicity test and analysis results showed CdSe QDs incubation caused pericardial edema, loss of vessels, decreased neuron expression and triggered intense in ammatory response. In order to reduce the side effects of nanomaterials, we choose some common modi cation methods. After CdSe QDs was encapsulated by silicon dioxide and the core-shell structure of CdSe@ZnS, the biocompatibility of CdSe QDs was signi cantly improved. Among these means of modi cation, the most biocompatible modi cation is polyethylene glycol (PEG) cladding method. In order to explore the improvement of CdSe QDs toxicity after PEG camou age, a series of transgenic zebra sh embryos were employed to evaluate the toxicity of CdSe QDs from these aspects of heart, blood vessel, immune system and nervous system. The results manifested CdSe@PEG could signi cantly increase the survival and hatching rate of zebra sh embryos as well as reduce the malformation rate compared with no camou age group. PEGcamou aged CdSe QDs was a more effective method to minimize the toxicity including the negative effects on the heart, nerves, blood vessels and in ammation. These nanomaterials in our work are randomly selected, and with this scanning method for embryo safety, any material of interest can be replaced, such as 2D metal carbides and nitrides (MXenes) and metal-organic frameworks (MOFs). We will continue to study the impact of nanomaterials on other tissue and organs, and further try to exchange for different polymer modi cations for screening and will study the mechanism of the in uence of nanomaterials on each system. Our work expects to inspire the more comprehensive evaluation strategies of biocompatibility for biomedical nanoparticles, especially in nerves, heart and blood vessels of embryos that are particularly sensitive to harmful substances.

Materials And Methods
Zebra sh lines and maintenance. Unless otherwise stated, the wild-type (Tu strain) embryos were used for experiments. The transgenic lines of Tg (kdrl: mcherry), Tg(coronin1a: EGFP) and Tg(elavl3:Gal4vp16,UAS:mcherry) line crossed with wild type respectively, and the transgenic embryos are selected for analysis of the vascular toxicity, in ammatory toxicity and the neurotoxicity. Tg (kdrl:mCherry) 29 were obtained from Dr. Bo Zhang (School of Life Science, Peking University, Beijing, China). The Tg (coronin1a: EGFP) line was provided by Dr Zilong Wen (Hong Kong University of Science and Technology, Hong Kong, China) 30 and Tg(elavl3: Gal4vp16,UAS:mcherry) line were kindly provided by Dr. Jiulin Du (Institute of Neuroscience, University of Chinese Academy of Sciences, Shanghai, China). All protocols for animal procedures were approved by the Animal Ethics Committee of Tianjin University. Different nanomaterial treatment. Select normal development embryos at 4 hpf under a stereomicroscope and place them in a 12-well plate (20 embryos in a well). The embryos were divided into 6 groups and one group with 3 parallel wells. The embryos were incubated in E3 culture medium with different concentrations of Au NPs, UCNPs, C QDs, ZnO QDs and CdSe QDs from 4 hpf to 72 hpf. In the nanomaterial modi cation screening experiment, the embryos were incubated in E3 culture medium with 0.1 mg/ml SiO 2 -modi ed CdSe QDs, ZnS-modi ed CdSe QDs and PEG-modi ed of CdSe QDs from 4 hpf to 72 hpf.. The mortality and deformity rates were recorded every day. Embryos hatching rate, body bending rate, pericardial edema rate were analyzed after nanomaterial treatment at 3 dpf.
In vivo study the effect of CdSe QDs and PEG-modi ed of CdSe QDs on heart. The wild type embryos were incubated in E3 culture medium and E3 with CdSe QDs and PEG-modi ed of CdSe QDs from 4 hpf to 72 hpf. After the removal of nanomaterials, the pericardial edema and heart rate were imaged by Nikon Ti2-U.
In vivo study the effect of CdSe QDs and PEG-modi ed of CdSe QDs on vascular. The heterozygote Tg (kdrl: mcherry) embryos were incubated in E3 culture medium and E3 with CdSe QDs and PEG-modi ed of CdSe QDs from 4 hpf to 72 hpf. After the removal of nanomaterials, the expression of vascular endothelial cells in control, CdSe QDs and CdSe@PEG group were imaged respectively by Leica THUNDER Imager Model Organism. The uorescence intensity was analyzed by LAS X software. The vessel branch was analyzed with Image J Angiogenesis Analyze.
In vivo study the effect of CdSe QDs and PEG-modi ed of CdSe QDs on in ammatory response. The heterozygote Tg (coronin1a: EGFP) embryos were incubated in E3 culture medium and E3 with CdSe QDs and PEG-modi ed of CdSe QDs from 4 hpf to 72 hpf. After the removal of nanomaterials, the expression of macrophages and neutrophils in control, CdSe QDs and CdSe@PEG group were imaged respectively by Leica THUNDER Imager Model Organism. The uorescence intensity was analyzed by LAS X software.
In vivo study the effect of CdSe QDs and PEG-modi ed of CdSe QDs on nerve.The heterozygote Tg (elavl3: mCherry) embryos were incubated in E3 culture medium and E3 with CdSe QDs and PEG-modi ed of CdSe QDs from 4 hpf to 72 hpf. After the removal of nanomaterials, the expression of neurons in control, CdSe QDs and CdSe@PEG group were imaged respectively by Leica THUNDER Imager Model Organism. The uorescence intensity was analyzed by LAS X software.
Zebra sh larvae behavioral analysis.The wild type embryos were incubated in E3 culture medium and E3 with CdSe QDs and PEG-modi ed of CdSe QDs from 4 hpf to 72 hpf. After the removal of nanomaterials, the embryos were place in the 24-well plate for behavioral analysis with DanioVision.
Statistical Analysis. Data were expressed as mean ± standard deviation (SD) of experiments. Data analysis was performed using Origin 8.0. The signi cance between groups was analyzed using an unpaired two-tailed t test (comparing two groups) and one-way analysis of variance (ANOVA) (comparing multiple groups) by Statistics Analysis System (*P < 0.05, **P < 0.01 and ***P < 0.001). P < 0.05 was considered signi cant.

Declarations
Ethical Approval and Consent to participate All animal experiments are approved by Tianjin University and consent to participate.

Consent for publication
Not applicable.

Availability of supporting data
All supporting data is available.

Competing interests
The authors declare that they have no competing interests.
Authors' contributions 1 The same contribution.