Developmental roles of tyrosine metabolism enzymes in the blood-sucking insect Rhodnius prolixus

Abstract

The phenylalanine/tyrosine degradation pathway is frequently described as a catabolic pathway that funnels aromatic amino acids into citric acid cycle intermediates. Previously, we demonstrated that the accumulation of tyrosine generated during the hydrolysis of blood meal proteins in Rhodnius prolixus is potentially toxic, a harmful outcome that is prevented by the action of the first two enzymes in the tyrosine degradation pathway. In this work, we further evaluated the relevance of all other enzymes involved in phenylalanine/tyrosine metabolism in the physiology of this insect. The knockdown of most of these enzymes produced a wide spectrum of distinct phenotypes associated with reproduction, development and nymph survival, demonstrating a highly pleiotropic role of tyrosine metabolism. The phenotypes obtained for two of these enzymes, homogentisate dioxygenase and fumarylacetoacetase, have never before been described in any arthropod. To our knowledge, this report is the first comprehensive gene-silencing analysis of an amino acid metabolism pathway in insects. Amino acid metabolism is exceptionally important in haematophagous arthropods due to their particular feeding behaviour.

1. Introduction

During feeding, haematophagous insects ingest many times their body weight in blood. As the protein content of vertebrate blood is very high, its digestion in the midgut generates a large flux of free amino acids. Therefore, tight regulation of amino acid metabolism, particularly tyrosine, is crucial for oogenesis [1], innate immune response [2] and survival [3]. Animals can obtain tyrosine from the hydroxylation of phenylalanine or from the hydrolysis of food proteins. Tyrosine can be used for the biosynthesis of proteins, biogenic amines and melanins or catabolized to energy by five enzymatic reactions resulting in acetoacetate and fumarate (figure 1a), which can be further catabolized through the Krebs cycle. The tyrosine catabolism pathway exists in eukaryotes and prokaryotes, and some human genetic diseases reflect the dysfunction of enzymes in this pathway. The clinical features of these disorders differ due to the accumulation or deficiency of different metabolites [4–6]. It has long been known that insects rely heavily on tyrosine for cuticle hardening and for the melanization of pathogens as a key component of the insect immune system; consequently, they possess more tyrosine metabolism enzymes than mammals [7]. In several insects, tyrosine storage has been described as the accumulation of tyrosine-rich proteins in the haemolymph [8] and as the formation of tyrosine-loaded vacuoles in the fat body [9].

Figure 1.

Several distinct phenotypes are produced upon silencing of different tyrosine metabolism enzymes. (a) Tyrosine metabolism pathways. PAH, phenylalanine hydroxylase; TAT, tyrosine aminotransferase; HPPD, 4-hydroxyphenylpyruvate dioxygenase; HgD, homogentisate 1,2-dioxygenase; MAAI, maleylacetoacetate isomerase; FAH, fumarylacetoacetase; TH, tyrosine hydroxylase; DDC, l-DOPA decarboxylase/aromatic l-amino acid decarboxylase; PO, phenoloxidase; DCT, dopachrome tautomerase; DCE, dopachrome conversion enzyme; TβH, tyramine β-hydroxylase. Metabolites are abbreviated as follows: l-DOPA (l-3,4-dihydroxyphenylalanine), DHI (5,6-dihydroxyindole), DHICA (5,6-dihydroxyindole-2-carboxylic acid). The image was modified from Sterkel et al. [3]. (b) The number of eggs laid by R. prolixus females in which enzymes of the Phe/Tyr degradation pathway and TH were silenced by dsRNA injection. (c) Hatching rate. (d) Survival of first-stage nymphs that hatched from eggs laid by females injected with dsRNA. (e) Fourth-stage nymph survival after injection with dsRNA. Cyan dots represent insects that died as fifth-stage nymphs, red dots represent insects that died as fourth-stage nymphs and blue dots represent insects that died during the ecdysis process. Dotted horizontal lines show the ecdysis period. At least two independent experiments were performed for each target gene, each with n = 8–12 insects per experimental group. The data from multiple experiments were combined into a single graph. The data were plotted as the mean ± s.e.m. Two-way ANOVA was performed to evaluate differences between the experimental and control groups in (a,b). The log-rank (Kaplan–Meier) test was used to evaluate significant differences in survival between the experimental and control groups in (d,e) (****p < 0.0001).

Rhodnius prolixus is one of the vectors of Trypanosoma cruzi, the parasite that causes Chagas disease, which affects approximately seven million people mainly in Central and South America [10]. Recently, its genome has been sequenced [11], which has allowed a systematic study of tyrosine metabolism enzymes. Previously, we showed that the inhibition of either tyrosine aminotransferase (TAT) or 4-hydroxyphenylpyruvate dioxygenase (HPPD)—the first two enzymes of the tyrosine catabolism pathway—causes the death of R. prolixus and other blood-feeding arthropods but is harmless to non-haematophagous insects, revealing an essential role of this pathway in the adaptation to haematophagy by detoxifying excess dietary tyrosine [3]. In this work, we further evaluated the relevance of all other enzymes involved in tyrosine metabolism in the physiology of R. prolixus. A wide spectrum of distinct phenotypes was obtained after the knockdown of these enzymes, with the only exception of maleylacetoacetate isomerase (MAAI), highlighting a pleiotropic role of tyrosine metabolism.

2. Material and methods

The same protocols described by Sterkel et al. [3] were used to perform this work.

(a). Rhodnius prolixus colony

Rhodnius prolixus were maintained under a photoperiod of 12 L : 12 D, at 28°C and 80–90% relative humidity. Insects were fed on rabbit at three-week intervals. Only adult mated females that had been fed once during the adult stage were used to perform the experiments.

(b). Synthesis of double-stranded RNA

Specific primers were designed to amplify each target gene by PCR (table 1). These primers contained the T7 polymerase binding sequence at 5′ end, which is required for double-stranded RNA (dsRNA) synthesis. The maltose-binding protein (MAL) gene from Escherichia coli (gene identifier 7129408) was used as a control for the off-target effects of dsRNA. MAL was amplified from the Litmus 28i-mal plasmid (New England Biolabs) using T7 minimal promoter primers. DsRNAs were synthesized using a MEGAscript RNA of interference (RNAi) kit (Ambion, Austin, TX), according to the manufacturer's instructions. dsRNAs concentrations were determined spectrophotometrically at 260 nm on a Nanodrop 1000 spectrophotometer v. 3.7 (Thermo Fisher Scientific) and visualized on an agarose gel (1.5% p/v) to confer dsRNA size, integrity and purity.

Table 1.

Sequences of the primers used to amplify target genes for RNA of interference experiments. T7 promoter sequences that were necessary for transcription are shown in red.

(c). RNA of interference to determine loss-of-function phenotypes

Rhodnius prolixus females and fourth-stage nymphs were injected with 2.5 µg of target gene dsRNA dissolved in 1 µl of sterile ultrapure water. Control insects were injected with 2.5 µg of MAL dsRNA. Insects were fed 7 days after the dsRNA injection, which was considered day 0 of the experiments. On that day, the efficacy of gene knockdown was checked by real-time PCR.

(d). RNA isolation, cDNA synthesis and quantitative real-time PCR

Rhodnius prolixus tissues were dissected in ice-cold PBS 7 days after dsRNA injection. Total RNA was extracted from tissues using TRIzol (Ambiol, Carlsbad, CA) according to the manufacturer's instructions. Following treatment with RNase-free DNaseI (Fermentas International, Burlington, Canada), first-strand cDNA synthesis was performed using 1 µg total RNA and a high-capacity cDNA reverse transcription kit (Applied Biosystems, Foster City, CA) and random hexamers, according to the manufacturer's instructions.

Specific primers for each gene were designed to amplify a different region from that amplified by the RNAi primers to prevent dsRNA amplification. They were also designed in different exons to prevent DNA amplification, and their efficiency was experimentally tested (table 2). The ribosomal protein 18S (rp18S) gene was used as a reference gene [12,13]. Quantitative PCR was performed using a StepOnePlus real-time PCR system (Applied Biosystems) using SYBR Green PCR Master Mix (Applied Biosystems) under the following conditions: one cycle for 10 min at 95°C, followed by 50 cycles of 15 s at 95°C and 45 s at 60°C.

Table 2.

Sequences of the primers used to amplify target genes by real-time PCR.

gene	GenBank accession number	forward primer	reverse primer	primer efficiency
phenylalanine hydroxylase	KX572140	GAGGTTGGTGCTTTGGTCAG	AGCTTCAAGGGCAGCATCTA	88.4
tyrosine aminotransferase	KX572141	TCATGCCACAAGGAGCTATG	TATACATGGAGGCGGACATTC	98.7
hydroxyphenylpyruvate dioxygenase	KX572142	GCTAAACAGGCGGCCAGCTA	TGGACGCTCTGTAACCAGGA	98.6
homogentisate 1,2-dioxygenase	KX572143	ACAGCATTTACTGCTCCTCG	CTGGAAATAGGCCTTCCTGC	87.6
maleylacetoacetate isomerase	KX572144	GACCTCAGCGACCTCTCCTA	ATCCAGTGCTGAGCCCATTC	91.45
fumarylacetoacetase	KX572145	TTGGCAGCGCCACTGTCGTA	TGGCTTCCAACCAACTTGCTG	98.9
tyrosine hydroxylase	KX572146	TAGCAGCAGCCCAGAAGAAC	GCTTAACGACGACGCTCTCA	75.3
ribosomal protein 18S	AJ421962	TGTCGGTGTAACTGGCATGT	TCGGCCAACAAAAGTACACA	89.2

(e). Tyrosine supplementation

Five microlitres of tyrosine (Sigma) 10 mM dissolved in sterile ultrapure water were injected on day 1 before a blood meal (BBM) and on day 2 PBM to PAH knockdown and control insects (dsMAL-injected insects).

(f). Carbidopa administration

Five microlitres of carbidopa (Sigma) 2.5 mM dissolved in H₂O were injected into the thorax at day 1 BBM and day 2 PBM, or at day 2 PBM and day 5 PBM at a late stage of oogenesis. Five microlitres of H₂O were injected into the control insects.

(g). Oviposition and eclosion ratios

After a blood meal, the females were individually separated into vials and kept at 28°C and 80–90% relative humidity, under a photoperiod of 12 L : 12 D. The number of eggs laid by each female was counted daily. The eclosion ratios were calculated by dividing the number of hatched first instar nymphs by the number of eggs laid by each female.

(h). Statistical analysis

All experiments were repeated at least twice. Statistical analyses were performed using Prism v. 6.0 software (GraphPad Software, San Diego, CA).

3. Results

Using gene information available from genome sequencing [11], RNAi was used to evaluate the effects on the physiology and development of R. prolixus of tyrosine metabolism enzymes. These enzymes either are involved in the production of tyrosine from phenylalanine or participate in pathways that use tyrosine as a substrate, and its degradation pathway that ultimately funnels tyrosine carbon skeletons into the Krebs cycle (figure 1a). As the R. prolixus genome contains four paralogous genes that encode the enzyme aromatic-l-amino acid decarboxylase (DDC), the enzyme that converts tyrosine into tyramine and DOPA into dopamine, the chemical inhibitor carbidopa was used to evaluate their physiological relevance. Intrathoracic injections of dsRNA drastically reduced the mRNA level of target genes (electronic supplementary material, figure S1). The knockdown of these enzymes, with the exception of MAAI, drastically negatively affected the reproductive fitness, moulting or survival of R. prolixus (figure 1b–e), further demonstrating the importance of tyrosine metabolism regulation.

The knockdown of phenylalanine hydroxylase (PAH), the enzyme that converts phenylalanine into tyrosine, affected the reproductive fitness of R. prolixus due to a combined effect of reducing the number of eggs laid per female and decreasing the egg hatching rate (figure 1b,c). Eggs did not develop into viable embryos, apparently becoming dehydrated (figure 2b), with an appearance resembling the phenotype described for the silencing of a Duox type NADPH oxidase, which was explained by the reduced cross-linking of eggshell proteins through di-tyrosine bonding, leading to defective eggshell waterproofing [14]. A similar phenotype affecting reproductive fitness after PAH silencing has been described in Aedes aegypti, associated with retarded vitellogenesis, a reduction in oviposition and impaired melanization of the chorion that resulted in a reduction in the hatching rate [1]. In that report, mosquito PAH silencing resulted in the accumulation of phenylalanine and other phenyl ketones, with a concomitant reduction in tyrosine levels. Because the DDC inhibitor carbidopa produced similar effects to dsPAH, these authors hypothesized that the phenotype was due to reduced levels of tyrosine available for dopamine synthesis [1]. By contrast, in R. prolixus, PAH-silenced insects accumulated phenylalanine in the haemolymph, but tyrosine levels were not reduced compared with control insects, indicating that tyrosine obtained from blood meal was sufficient to maintain physiological tyrosine levels without the requirement of phenylalanine hydroxylation [3]. Similarly, the supplementation of exogenous tyrosine failed to revert or alleviate the phenotype (figure 3). Furthermore, silencing of tyrosine hydroxylase (TH) (figure 1b,c) and inhibition of DDC by carbidopa (figure 4) did not reproduce the PAH gene-silencing phenotype, suggesting that the phenotype observed was not due to a deficiency of tyrosine or any other downstream metabolite, such as DOPA, dopamine or tyramine. Taken together, these results indicate that the phenotype observed after PAH silencing was caused by the accumulation of phenylalanine (or another derived metabolite). It is also possible that at least part of the physiological impact observed was not caused by the decreased PAH enzymatic activity, as the 4-α-carbinolamine dehydratase subunit of this enzyme is a dual-function protein that, in addition to its function in phenylalanine metabolism, also works as a co-activating transcription factor [15].

Figure 2.

Silencing of PAH, FAH and TH interferes with embryo development and hatching. (a) Control egg (dsMAL). (b) Egg from female injected with dsPAH. (c) Egg from female injected with dsFAH. (d) Egg from female injected with dsTH. Images were taken 20–21 days after the eggs were laid.

Figure 3.

Supplementation with exogenous tyrosine to R. prolixus females does not rescue or alleviate the PAH silencing phenotype. Control (MAL) and PAH-silenced R. prolixus females received two injections of 50 nmol of tyrosine at day 1 before blood meal and at day 2 PBM. At least two independent experiments were performed for each target gene. The data from multiple experiments were combined into a single graph. The data were plotted as the mean ± s.e.m. Two-way ANOVA was performed to evaluate differences between the experimental and control groups in (a,b). The log-rank (Kaplan–Meier) test was used to evaluate significant differences between the experimental and control groups in panel (c) (****p < 0.0001).

Figure 4.

Administration of l-DOPA decarboxylase/aromatic l-amino acid decarboxylase (DDC) inhibitor carbidopa to R. prolixus females delays oviposition and decreases the hatching of embryos. (a) Oviposition. (b) Egg hatching. (c) First-stage nymph survival. Five microlitres of carbidopa 2.5 mM dissolved in H₂O were injected into the thorax on day 1 BBM and day 2 PBM, or on day 2 PBM and day 5 PBM. Two independent experiments were performed for each carbidopa treatment, each with n = 8–12 insects per experimental group. The data from both experiments were combined into a single graph. The data were plotted as the mean ± s.e.m. Two-way ANOVA was performed to evaluate differences between the experimental and control groups in (a,b). The log-rank (Kaplan–Meier) test was used to evaluate significant differences between the experimental and control groups in (c) (**p < 0.01).

A drastic decrease in the hatching rate was also observed in insects injected with dsTH (figure 1c), but, unlike insects treated with dsPAH, embryo development proceeded until the late stages of embryogenesis (figure 2d). Nevertheless, most first-stage nymphs were unable to hatch. TH knockdown also resulted in a decrease in cuticle hardness and drastically affected the ecdysis process in fourth-stage nymphs, and all of the insects died during moulting (figure 1e). A critical function of TH during hatching has also been described in holometabolous (non-haematophagous) insects. In Bombyx mori, TH dysfunction prevented pigmentation and hatching [16]. Upon microscopic inspection of the embryos, it was observed that unhatched embryos almost completed development and attempted to hatch, but failed to break the eggshell with their mouthparts to emerge. Feeding l-dopa to neonate larvae rescued the phenotype. The authors concluded that lower amounts of TH downstream products such as DOPA, dopamine and/or melanin resulted in insufficient hardening of the larval mandibles to allow eggshell disruption through biting [16]. In Tribolium castaneum, a reduction in TH function mediated by RNA interference resulted in a decrease in cuticle pigmentation and sclerotization [17]. In Drosophila melanogaster, TH null mutants resulted in unpigmented embryos that were also unable to hatch [18]. Here, we describe for the first time a similar phenotype affecting the hatching of embryos and the ecdysis of nymphs in a hemimetabolous insect, R. prolixus, demonstrating that the function of this enzyme is conserved among insects.

The R. prolixus genome encodes four distinct DDC genes [11]. Their inhibition by carbidopa injections resulted in a small reduction in the hatching rate of eggs laid by injected females (figure 4b). As carbidopa can also inhibit the biosynthesis of serotonin from 5-hydroxy-l-tryptophan [19], the results presented here may indicate the involvement of the tyrosine-derived neurotransmitters tyramine and dopamine or serotonin in the regulation of R. prolixus reproductive fitness.

In humans, the most severe disease associated with amino acid metabolism is hereditary tyrosinaemia type 1 (HT1), which is caused by a genetic deficiency of FAH. This genetic disorder is characterized by progressive liver disease and renal tubular dysfunction if left untreated, and is due to the accumulation of fumarylacetoacetate (FAA), maleylacetoacetate (MAA) and succinylacetone (SAA) [5]. FAH silencing in R. prolixus did not affect insect survival [3], but we show here that its silencing compromised the late stages of embryogenesis and the hatching of nymphs (figure 1c). The few first-stage nymphs that hatched presented deformed abdomens and legs, and in many cases their legs were trapped in the embryonic membrane. As a consequence, those nymphs that managed to hatch from eggs died soon after eclosion (figure 5). Thus, FAH knockdown in adult females produced the complete suppression of reproduction.

Figure 5.

Silencing of FAH affects embryo development. Shown are phenotypes observed in FAH-silenced first-stage nymphs (N1) that hatched from eggs laid by females injected with dsFAH, revealing an important role of this enzyme in embryo morphogenetic events. (a) Dorsal view of control N1. (b) Ventral view of control N1. (c) Dorsal view of FAH-silenced N1. (d) Ventral view of FAH-silenced N1.

First-stage nymphs hatched from eggs laid by females injected with dsPAH, dsHgD, dsMAAI or dsTH showed apparently normal morphology. However, survival was strongly reduced for nymphs that hatched from eggs laid by females injected with dsHgD (figure 1d). The knockdown of homogentisate dioxygenase (HgD) also caused the death of fourth-stage nymphs (figure 1e), but, unlike dsTH-injected nymphs, they moulted normally and died a few days later. The dysfunction of HgD in humans causes a disease called alkaptonuria. This disorder is due to the inability of the body to deal with homogentisic acid [5,6]. In R. prolixus, HgD knockdown did not affect the adult female lifespan [3], reproductive fitness or ecdysis, but it reduced the lifespan of the nymphs (figure 1d,e). Further studies are required to determine the molecular mechanism causing nymph death.

Recent work in Caenorhabditis elegans demonstrated that tyrosine is a metabolic signal that influences developmental decisions and longevity. Tyrosine accumulation due to the knockout of TAT promoted the arrest of larval development at the dauer stage and even extended its lifespan [20]. In contrast to C. elegans, the knockdown of TAT and HPPD in R. prolixus fourth-stage nymphs reduced their survival, presenting the same lethal phenotype previously observed in adult insects, associated with tyrosine accumulation [3]. However, similar to the phenotype described in C. elegans, tyrosine accumulation prevented the normal ecdysis process, and many insects did not moult, despite their survival being longer than the ecdysis period, indicating that tyrosine also influences developmental decisions in R. prolixus (figure 1e).

MAAI is also known as glutathione transferase zeta (GSTZ1) [21]. In addition to its function in tyrosine degradation, it is involved in the detoxification of xenobiotic compounds and is able to catalyse the conversion of dichloroacetic acid to glyoxylic acid [22,23]. In humans, MAAI is the only enzyme in the tyrosine catabolism pathway for which failure does not produce a disease. MAAI-deficient mice accumulate FAA and SAA in urine but otherwise appear healthy. Mice that are double-mutant for MAAI and FAH die rapidly on a normal diet, as do FAH mutants, indicating that MAA can be isomerized to FAA in the absence of MAAI, possibly by a glutathione-dependent non-enzymatic bypass mechanism [24], which may also explain the lack of a visible phenotype in MAAI-silenced R. prolixus.

4. Discussion

Previously, we showed that tyrosine detoxification by TAT and HPPD, the first two enzymes in the tyrosine oxidative degradation pathway, prevents the accumulation of high and lethal levels of tyrosine, constituting an essential metabolic adaptation to haematophagy [3]. In this work, we assessed the effect of silencing of all enzymes of the tyrosine degradation pathway and of key enzymes of all other metabolic pathways related to tyrosine metabolism. The markedly distinct phenotypes obtained when different enzymes were silenced may reflect the accumulation or lack of different metabolites, and indicate that different mechanisms are involved in each case. The knockdown of most of these enzymes drastically affected fundamental processes of R. prolixus physiology, such as reproduction, embryogenesis, ecdysis and nymph survival, further highlighting the particular importance of tyrosine metabolism in R. prolixus physiology. The highly pleiotropic consequences of the silencing of these genes provide a view of the tyrosine oxidation pathway that suggests a much higher connectivity than the traditional view that describes this pathway as a simple tyrosine mill that directs fragments of the amino acid carbon skeletons into the Krebs cycle. Particularly important are the possible links to developmental regulatory pathways, a possibility that deserves future mechanistic research. To our knowledge, this report is the first comprehensive gene-silencing analysis of an amino acid metabolism pathway in insects. Amino acid metabolism is particularly important in haematophagous arthropods, as proteins account for more than 80% of vertebrate blood dry-weight. This area is poorly studied, and follow-up studies may lead to the identification of new targets for the development of alternative strategies for the control of vector populations and the diseases they transmit.

Ethics

Dedicated technicians carried out all aspects related to rabbit husbandry under strict guidelines. All animal care and experimental protocols were conducted according to the guidelines of the institutional care and use committee (Committee for Evaluation of Animal Use for Research from the Federal University of Rio de Janeiro) and the National Institute of Health Guide for the Care and Use of Laboratory Animals (ISBN 0-309-05377-3). The protocols received registry number 115/13 from the Animal Ethics Committee (Comissão de Ética no Uso de Animais, CEUA).

Authors' contributions

M.S. designed and performed the experiments, analysed the data and wrote the paper. P.L.O. designed the experiments, analysed the data and wrote the paper. Both authors discussed the results and contributed to the final version of the manuscript.

Competing interests

We declare we have no competing interests.

Funding

This research was funded by CAPES, FAPERJ, CNPq and INCT.