Evaluation of the silkworm lemon mutant as an invertebrate animal model for human sepiapterin reductase deficiency

Human sepiapterin reductase (SR) deficiency is an inherited disease caused by SPR gene mutations and is a monoamine neurotransmitter disorder. Here, we investigated whether the silkworm lemon mutant could serve as a model of SR deficiency. A point mutation in the BmSPR gene led to a five amino acid deletion at the carboxyl terminus in the lemon mutant. In addition, classical phenotypes seen in SR deficient patients were observed in the lemon mutant, including a normal phenylalanine level, a decreased dopamine and serotonin content, and an increased neopterin level. A recovery test showed that the replenishment of l-dopa significantly increased the dopamine level in the lemon mutant. The silkworm lemon mutant also showed negative behavioural abilities. These results suggest that the silkworm lemon mutant has an appropriate genetic basis and meets the biochemical requirements to be a model of SR deficiency. Thus, the silkworm lemon mutant can serve as a candidate animal model of SR deficiency, which may be helpful in facilitating accurate diagnosis and effective treatment options of SR deficiency.

FD, 0000-0002-0215-2177 Human sepiapterin reductase (SR) deficiency is an inherited disease caused by SPR gene mutations and is a monoamine neurotransmitter disorder. Here, we investigated whether the silkworm lemon mutant could serve as a model of SR deficiency. A point mutation in the BmSPR gene led to a five amino acid deletion at the carboxyl terminus in the lemon mutant. In addition, classical phenotypes seen in SR deficient patients were observed in the lemon mutant, including a normal phenylalanine level, a decreased dopamine and serotonin content, and an increased neopterin level. A recovery test showed that the replenishment of L-dopa significantly increased the dopamine level in the lemon mutant. The silkworm lemon mutant also showed negative behavioural abilities. These results suggest that the silkworm lemon mutant has an appropriate genetic basis and meets the biochemical requirements to be a model of SR deficiency. Thus, the silkworm lemon mutant can serve as a candidate animal model of SR deficiency, which may be helpful in facilitating accurate diagnosis and effective treatment options of SR deficiency. of the individual, because diseases reflect the combined action of genes and environment. Often, it is too dangerous to carry out studies in patients directly and doing so involves ethical issues. To solve these problems, animal models of human disease are used in the laboratory. Animal models of human disease have many advantages. For example, there is no spatio-temporal limit. It is easy to research human diseases that have long duration and low incidence. Researchers also can study the effects of various conditions on the disease. To date, there are more than 5000 animal models of human diseases. Common species used as animal models of human diseases include monkey (Primates) [1], mouse (Mus musculus) [2], zebrafish (Danio rerio) [3], fruit fly (Drosophila melanogaster) [4] and nematode (Caenorhabditis elegans) [5]. The main diseases studied include acquired immune deficiency syndrome, cancer, diabetes, nervous system diseases and cardiovascular diseases. These animal models have been chosen for their respective advantages and characteristics. However, they still have shortcomings, such as limited resources, expensive experimental costs, long experimental periods and animal ethics and welfare disputes. Silkworms (Bombyx mori) are not only economic insects, but have also been useful model animals in genetics for more than a century. Additionally, the silkworm has 8469 human homologous genes with 58% homology [6].
The silkworm has many advantages as an animal model of human disease, such as abundant mutant resources, a clear genetic background, a short life cycle, a high fertility rate, a moderate body size, convenient dissection, easy operation for haemolymph and midgut injection, as well as no animal ethical controversy. Until now, there have been some reports on the use of silkworms to construct disease models, such as a hyperlipidemic model [7], a hyperproteinemic model and a hyperglycemic model [8,9]. In addition, there are silkworm mutants that have the potential to be used as the models of human diseases. For example, the silkworm oa mutant is caused by a mutation in the BmHPS5 gene, which is homologous with the Hermansky-Pudluck syndrome-5 (HPS5) gene in humans [10]. Down-regulation of the DJ-I gene in the silkworm op mutant affects the plasma uric acid synthesismodulating pathway and results in a similar phenotype to the clinical features of Parkinson disease [11]. The pale body colour silkworm (albino, al) is caused by mutation of the BmPTPS gene, and the results of related gene expression analysis and recovery experiments indicated that the mutant might be a potential animal model of tetrahydrobiopterin (BH4)-deficient phenylketonuria [12,13]. The silkworm also plays a role in drug screening, and there have been pathogenic bacterial infection models [14], pathogenic fungal infection models [15], as well as antibiotic drug screening models [16]. Therefore, current studies provide an important theoretical basis for using the silkworm to analyse the pathogenesis of human diseases and develop therapeutic drugs.
In 2001, SR deficiency (OMIM: 612716), another type of BH4 deficiency without hyperphenylalaninemia, was fully recognized [30]. SR deficiency is an inherited autosomal recessive disorder [31], and its incidence is unknown. To date, 13 different mutations have been found in 44 SR deficient patients (http://www.biopku.org) [32]. SR deficiency is a neurotransmitter disorder caused by SPR gene (2p14-p12) mutations [31]. Its core clinical features mainly include cognitive problems, such as royalsocietypublishing.org/journal/rsos R. Soc. open sci. 7: 191888 mental retardation, extreme mood swings and language delay [33], and motor problems including dystonia, axial hypotonia, muscle stiffness and tremors, seizures and oculogyric crises [34,35]. SR deficiency cannot be detected by newborn screening because it is not companied by hyperphenylalaninemia [36] and initially is often misdiagnosed as other diseases, such as DHPR deficiency [37]. This results in physical damage owing to delayed treatment, which should be initiated as early as possible. The exact diagnosis must be made with mutation analysis of the SPR gene combined with an analysis of pterins and biogenic amines in cerebrospinal fluid [38]. SR deficiency displays decreased levels of homovanillic acid (HVA) and 5-hyroxyindolacetic acid (5-HIAA), two products of neurotransmitters, and elevated total biopterin (tBio), dihydrobiopterin (BH2) and sepiapterin, as well as normal to slightly increased levels of neopterin. The treatment of SR deficiency aims to correct central nervous system (CNS) dopamine deficiency, so the common method is to supply L-dopa, the precursor of dopamine, and carbidopa, a peripheral decarboxylase inhibitor [39]. It is important to initiate treatment as early as possible to avoid irreversible neurological damage. Some challenges in SR deficiency remain. One is that newborn screening is difficult to obtain, but is key to prevent mental disability. Another problem is that the treatment method is too specific to meet the needs of all patients with different mutations, which brings great difficulties for clinical discovery and treatment of the disease. Thus, suitable animal models are needed to resolve these problems and to gain a better understanding of other similar human diseases.
In this study, we explored whether the lemon mutant could serve as an animal model of SR deficiency. Molecular cloning showed that the BmSPR gene of the lemon mutant had a point mutation. In addition, biochemical studies revealed a decreased level of dopamine and serotonin, an increased level of neopterin and a normal level of phenylalanine (LPA). Furthermore, the level of dopamine increased significantly after feeding L-dopa. Additionally, the lemon mutant showed negative behavioural abilities. Interestingly, unlike the results for human patients, the lemon mutant displayed a normal level of blood trehalose. Our findings indicate that the lemon mutant can serve as a candidate animal model of human SR deficiency, which will play an important role to uncover the pathogenic mechanism and for screening drugs for SR deficiency and other similar monoamine neurotransmitter diseases.

Silkworm strain and feeding conditions
All silkworm strains used in the experiments were obtained from the Silkworm Gene Bank at Southwest University. Their environment was 25°C, 75% relative humidity, and a 12 L : 12 D photoperiod. Mulberry leaves were their only food during the entire larval stage [40].

Cloning of the BmSPR gene
We downloaded the full-length sequence of the BmSPR gene from SilkDB (silkworm.genomics.org.cn) and designed primer pairs to target the whole coding sequence (CDS). After sequencing cDNA clones, we aligned the obtained sequence with the sequence submitted on NCBI. The primers used for polymerase chain reaction (PCR) are shown in table 1.

Reverse transcription-quantitative polymerase chain reaction
The silkworms were collected from day 1 of the first instar to day 1 of the fifth instar.

Measurement of body weight and length
The wild-type and lemon mutants at day 3 of fifth instar larvae were selected randomly and split into three groups to measure the body length and weight before feeding [40].  We designed a mulberry leaf-luring test and a turnover test to investigate the locomotor activity of the lemon mutant. In the mulberry leaf-luring test, silkworm lemon mutants and wild-type silkworms were chosen randomly 5 h after feeding. The two groups of silkworms were placed 0.5 cm away from the chronograph line, which is the pre-crawl distance. At 5 cm away from the chronograph line, we put mulberry leaves to induce silkworm crawling. When the silkworms climbed to the timing line, we began to count. The number of silkworms crawling to the mulberry leaves and the time required by each group were counted within 15 min. In the turnover test, the amount of time required for the silkworm to turn upside down was measured. Day 3 of fifth instar larvae of silkworm lemon mutants and wild-type silkworms were chosen for the two tests. Scatter plots of the two tests were drawn using the Column mode in GRAPHPAD PRISM5.

Serum phenylalanine and trehalose assays
The serum phenylalanine level was measured by a kit purchased from Suzhou Comin Biotechnology Co. Ltd (Suzhou, China). The serum trehalose level was measured using a kit from Beijing Suo Laibao Biotechnology Co. Ltd (Beijing, China). All steps followed the manufacturer's instructions.

Neopterin measurements
The ultra performance liquid chromatograph (UPLC) system (Shimadzu, Japan) used here consists of a SIL-30AC automatic sampler, an LC-30AD solvent delivery module, a CTO-30A column oven, a CBM

Dopamine and serotonin measurements
The silkworm brains were removed into a cold homogenizing buffer and blended using a JXFSTPRP-32 homogenizer for 240 s at 70 Hz. The homogenizing buffer contained 444.5 mg L-ascorbic acid and 61 ml hydrochloric acid (36%) with water to 500 ml. Afterward, the sample was centrifuged at 12 000 rpm for 20 min at 4°C. The supernatants were transferred into clean 1.5 ml microcentrifuge tubes in a boiling water bath for 3 min. After cooling, the same volume of dichloromethane was added to the supernatants and vortexed (PTR-60) for 10 min. After centrifuging the mixture at 12 000 rpm for 15 min, the supernatants were frozen at −80°C until use. We used a Shim-pack XR-ODS III column (1.6 µm, 2.0 mm internal diameter × 75 mm). The mobile phase was a mixture of 0.021 g 1-octanesulfonate monohydrate, 0.038 g disodium edetate dehydrate, 6.78 ml phosphoric acid and 100 ml methylalcohol. After adjusting to pH 2.68 by NaOH, the mobile phase was filtered using 0.22 µm filters. The separation was performed with a flow rate of 0.2 ml min −1 at 40°C. Dopamine and serotonin were detected at λ ex/em 280/330 nm. The UPLC system was the same as that used for neopterin measurements.

L-dopa feeding experiment
Each lemon mutant was fed with mulberry leaves with 4 mM, 6 mM and 8 mM L-dopa and carbidopa mixture (treatment group) or deionized water (control group) from day 1 to day 3 of fifth instar larvae. The dopamine level in the brains of the four groups was detected by UPLC.

Statistical analysis
GRAPHPAD PRISM 5 software (La Jolla, CA, USA) was used to analyse the statistical data. The results are shown as mean ± s.e.m. and the significance was determined with a two-tailed Student's t-test. Values of p < 0.05 were statistically significant.

Yellow body colour and BmSPR gene mutation in the silkworm lemon mutant
Silkworm lemon mutant, a spontaneous autosomal recessive mutant, has yellow body colour like the Japanese lem mutant [43], in contrast with the normal pigmentation of the wild-type (figure 2), which might be caused by abnormalities in the pigment metabolism pathway. Body colour pigments in insects mainly include ommochrome, pterin and melanin. Previous studies have reported that the yellow body coloration of the Japanese lem mutant is because of the accumulation of sepiapterin, yellow pteridines and sepialumazine in the integument [44]. In 2009, Meng et al. [43] demonstrated that the BmSPR gene was responsible for the Japanese lem mutant. Therefore, we investigated mutations in the BmSPR gene in the lemon mutant. We found that the BmSPR gene of the lemon mutant had a point mutation same to the change of the gene in the Japanese lem mutant (figure 3a), which leads to premature translation termination and a five amino acid deletion at the carboxyl terminus (figure 3b). Furthermore, we investigated the relationship between this yellow body colour and the BmSPR gene of the lemon mutant through a multi-line validation experiment. We found that the BmSPR gene of the nine yellow body colour strains had the same point mutation (figure 3c,d), indicating that the yellow body colour is related to the BmSPR gene. This phenotype suggests that the silkworm lemon mutant may be an appropriate animal model. These data showed that the lemon mutant has the genetic basis as an animal model of human SR deficiency, which is an inherited disease caused by SPR gene mutations.

Higher expression of the BmSPR gene at early larva instar in the silkworm lemon mutant
To understand the expression of the BmSPR gene in the lemon mutant, we investigated the temporal expression profile of BmSPR in the wild-type and lemon mutant. The results showed that the BmSPR gene is expressed in each period (from first instar to fifth instar) (figure 4a). Among them, the expression level of the BmSPR gene in the lemon mutant was higher during the first day of the second instar than in other periods, which indicated that the BmSPR gene is more important in the early stage of silkworm larvae. This is consistent with human SR deficiency, which shows clinical features in infancy. We also investigated the expression level of important genes in the BH 4 synthesis pathway, including BmGTPCH I, BmPTPS, BmDHFR, BmPAH and BmTH in the lemon mutant during the early stages ( figure 4b-f ). The results displayed that the expression changes of these five genes is reasonable when the BmSPR gene mutates, which also confirms the conservation of BH 4 synthesis and metabolic pathways in silkworm and humans.

Normal development and negative locomotion activities in the silkworm lemon mutant
The clinical manifestations of SR deficiency include stunting, dyskinesia and dystonia. Therefore, we measured the body weight and length of the silkworm lemon mutant. The results showed that the wild-type L5D3 wild-type L2D1 lemon L5D3 lemon L2D1 body length and weight of the lemon mutant and the wild-type silkworm were similar, meaning that the BmSPR mutation is not detrimental to body length and weight of the lemon mutant (figure 5a,b). We also measured the blood glucose level, but we did not investigate the difference between the silkworm lemon mutant and the wild-type (figure 5c). Then, we explored the locomotion ability of the lemon mutant through the mulberry leaf-luring test and the turnover test (figure 5d,e). We found that the lemon mutant needed more time to arrive the mulberry leaves than the control group; however, there was no significant difference in the time needed for the turnover test between the lemon mutant and the wild-type. According to these data, we believe that the BmSPR gene has a negative effect on the behavioural abilities of the lemon mutant.

Similar biochemical features of sepiapterin reductase deficiency in the lemon mutant
SR deficiency does not show hyperphenylalaninemia, which is one of its diagnostic features. Therefore, we compared the phenylalanine content in the blood of the lemon mutant and the wild-type. We found that the blood phenylalanine content was not significantly different between the lemon mutant and the wild-type (figure 6a). The dopamine and serotonin levels in the cerebrospinal fluid of the SR deficiency patients is significantly low, which is an important indicator for clinical diagnosis. We assessed the content of these two neurotransmitters in the brains of the lemon mutant and the wildtype by UPLC. We found that the levels of dopamine and serotonin in the lemon mutant were significantly lower than those in the wild-type ( figure 6b,c). SR deficiency shows an increased level of  royalsocietypublishing.org/journal/rsos R. Soc. open sci. 7: 191888 neopterin. We assayed the level of neopterin in the brains of the lemon mutant and the wild-type. We found that the content of neopterin in the lemon mutant was significantly higher than that in the wild-type (figure 6d). All data were consistent with the clinical features of SR deficiency; therefore, the lemon mutant meets the biochemical requirements as an animal model of SR deficiency.

Increased level of dopamine by oral supplementation of L-dopa in the lemon mutant
The main treatment for SR deficient patients is to correct the CNS dopamine deficiency by treatment with L-dopa and carbidopa. Therefore, we gave the lemon mutant three concentrations of these two compounds. The results showed that the dopamine level in the brains of silkworms in the three experimental groups was significantly higher than that in the control group, which is consistent with the treatment outcome of SR deficiency at the biochemical level ( figure 7). However, there was no significant change in the phenotype, which might be because the concentrations of L-dopa and carbidopa were too low to form enough dopamine for melanin synthesis.

Discussion
SR deficiency is a monoamine neurotransmitter disorder caused by abnormalities in the synthesis, transport and metabolism of monoamines. SR deficiency is difficult to diagnose accurately and lacks effective medicines for a diverse patient population, which misses the best cure time and causes nerve injury. Thus, animal models of SR deficiency are required for further studies. Meng et al. [43] proved the Japanese lem mutant is caused by the mutation of the BmSPR gene. There is a mutant called lemon whose body colour is similar to the Japanese lem's in the Southwest University. In this study, we aimed to explore whether the lemon mutant could serve as a model of SR deficiency. We found that the lemon mutant shows a BmSPR mutation, negative motor ability, normal phenylalanine level, decreased dopamine and serotonin levels, and increased neopterin level. These characteristics of the lemon mutant resemble the symptoms of SR deficiency in humans. Moreover, the administration of L-dopa and carbidopa increased the dopamine level of the lemon mutant. These findings support that the lemon mutant can be regarded as an animal model of SR deficiency in terms of genetic and biochemical aspects that are useful for studying precise diagnosis and for screening effective compounds.
BmSPR was highly expressed in the first and second instar larvae in the lemon mutant, suggesting that the gene is more important at an early stage in silkworms, which is consistent with the outcomes of SR deficiency in infancy in humans. In addition, the expression of other key genes in the BH 4 pathway suggests that it is conserved between silkworms and humans. However, we did not observe growth retardation, which is a classical clinical feature of SR deficient patients, in the lemon mutant. This might be because of species differences. The CDS of the human SPR gene is 786 bp, which encodes a 261 amino acid protein. Meanwhile, the CDS of the BmSPR gene is 801 bp, and the SPR enzyme is 266 amino acids long. Therefore, the function of BmSPR may differ between species. Administration of L-dopa is the main treatment for SR deficiency and aims to restore the dopamine level to rescue nerve damage. Although the dopamine level of the lemon mutant increased remarkably after feeding with L-dopa, we did not observe a significant phenotype change. In insects, dopamine is involved in the synthesis of melanin. Therefore, we suggest that the concentrations of L-dopa and carbidopa were too low to form enough dopamine for melanin synthesis, or the content of synthesized melanin was too low to show significant body colour changes.
The silkworm has many advantages and potential as an animal model of human diseases. With the completion of the genome map, as well as fine mapping and multi-strain re-sequencing for the silkworm, analysis of silkworm mutant mechanisms has improved. Rich silkworm mutant resources could potentially be used as disease models. The advantages of spontaneous animal models are that the occurrence and development of the diseases may be very similar to the corresponding diseases in humans. These silkworm mutants can not only help us understand the pathogenesis of various human diseases but also increase the possibility of discovering new genes for these diseases. There are 8469 human homologous genes in the silkworm with a 58% homology rate, which is close to the homology There are 5006 genes related to 1612 human diseases which have corresponding homologous groups in the silkworm. These diseases are classified into 18 categories. The corresponding homologous genes in the silkworm are mainly related to skeletal, head, neck, neurological and developmental diseases. Therefore, the silkworm is a very promising animal model of human diseases, which can play a crucial part in revealing disease mechanisms and drug screening.

Conclusion
In summary, the lemon mutant is the first invertebrate model of human SR deficiency. We believe it will be an important resource to address questions of better diagnostic criteria and effective therapies for SR deficiency and other neurotransmitter disorders.