Invasive anisakiasis by the parasite Anisakis pegreffii (Nematoda: Anisakidae): diagnosis by real-time PCR hydrolysis probe system and immunoblotting assay

Total genomic DNA was extracted from 2 mg of homogenized tissue from the two nematodes removed by endoscopy in CC1 and from one larva of CC2, by using the “DNeasy blood and tissue” kit (Qiagen), following the manufacture’s protocol.

In CC3, a single one histological section, obtained from the surgically removed granuloma at the intestinal level, was available. Thus, the larval nematode - embedded in the paraffin - was gently removed by scratching the histological slice from the glass-slide, by using a glass rod bearing a needle tip. After that, the parasite genomic DNA extraction was carried out according to procedures previously described [11].

The genomic DNA obtained from those samples are stored at the research laboratory of S. Mattiucci, Department of Public Health and Infectious Diseases (Section of Parasitology) - ‘Sapienza University of Rome’.

In CC1 and CC2, the identification at species level was previously obtained by direct sequences analysis of mitochondrial (mtDNA cox2, 629 bp) [18] and nuclear (elongation factor EF1 α-1 of nDNA, 409 bp) genes [19]. Whereas, the extremely low amount of DNA (<0.01 ng/μl, Qubit2.0 BR-dDNA kit) obtained from the single one histological slice available in CC3, did not allow its direct sequencing.

The mitochondrial cytochrome c oxidase subunit II (cox2) gene was amplified using the primers 211F (5′-TTTTCTAGTTATATAGATTGRTTYAT-3′) and 210R (5′-CACCAACTCTTAAAATTA TC-3′) [17]. Polymerase chain reaction (PCR) was carried out according to the procedures as previously described [18]. The sequences obtained at the mtDNA cox2 for the three larval nematodes, analyzed in the present study, were compared with those already obtained for the same gene in the species A. pegreffii [18].

The elongation factor (EF1 α − 1 nDNA) nuclear gene was amplified using the primers EF-F (5′-TCCTCAAGCGTTGTTATCTGTT-3′) and EF-R (5′-AGTTTTGCCACTAGCGGTTCC-3′) according to Mattiucci et al. [19]. The PCR conditions and procedures followed those reported in Mattiucci et al., 2016. The sequences obtained at the EF1 α − 1 gene of the nDNA for the three larval specimens and analyzed in the present study were compared with those already obtained for the same gene in the species A. pegreffii, as previously described [19].

The identification at species level of CC3 was obtained by using a RT-PCR species-specific primers/probe system [see below] based on mtDNA cox2 gene. The same method was applied in the DNA obtained from the larvae of CC1 and CC2. In addition, in order to verify the specificity of the RT-PCR primers/probe system and to test a possible cross-reactivity with the host tissue, the genomic DNA from human blood (h) was also tested in the same reaction. The RT-PCR primers/probe system was carried out using the primers and specific probe for A. pegreffii as follows: forward primer RTpegF-CTTTTGGAGGTTGATAATCG and reverse primer RTpegR-CCCACAAATCTCTGAACATT; species-specific probe pegHyPr-CTTGGGCTTTGCCTAGGATGTC (dual labeled with 6FAM/BHQ) [our unpublished data]. PCR amplification reaction was performed in a total volume of 15 μl, containing 7.5 μl of LC FS DNA HyProbe MASTER (2×) (Roche®), 0.3 μl of primers (0.2 μM), 0.75 μl of probe (0.3 μM) and 1.5 μl (3–5 ng/μl) of DNA template. Reactions were carried out as follows: 95 °C for 10 min (initial denaturation), followed by 40 cycles at 95 °C for 10 s (denaturation), 55–65 °C for 30 s (annealing), 72 °C for 1 s (extension), followed by a cooling step at 40 °C for 30 s. The reactions were run in a Light Cycler 480 II System (Roche®) Real Time PCR instrument, using LightCycler 480 Multiwell Plate-96 white (Roche®) and LightCycler 480 Sealing Foil (Roche®). Samples were tested in duplicate. The results were analyzed using the LC480 dedicated Software.

The parameters of efficiency and detection limit were estimated to test the performance of the assay. The efficiency of primers was tested; an amplification reaction was considered optimal if the efficiency value was 2 (E = 10^(−1/slope)); the E value was automatically calculated by the software program of the RT- Roche Instrument), because the amount of target DNA would double with each amplification cycle [20]. The Slope of the standard curve describes the kinetics of the PCR amplification and indicates how quickly the amount of target DNA can be expected to increase with the amplification cycles; in an optimal amplification reaction, the slope value was −3.3, in relation to the E value.

In addition, in order to establish the minimum amount of DNA detectable by the system, the Limit Of Detection (LOD) was established, preparing 16 serial 2-fold DNA dilutions starting from a known amount of DNA (from 20 ng/μl to 0.0006 ng/μl).

The immune response was tested against antigens/allergens Anisakis pegreffii larvae for: IgE and IgG₄ in CC1, IgE and IgG in CC2 and CC3. In particular, Excretory/Secretory (ES) antigens obtained from in vitro culture of A. pegreffii, were used in the Immunoblotting assay [21]; they followed the designation as reported in our previous paper [21].

Antigens from crude extracts and ES were run through protein gel electrophoresis. Proteins were diluted with dissociation buffer (1:1) and boiled at 95 °C for 5 min. Boiled proteins were run through 12% SDS-polyacrylamide gel and transferred on a nitrocellulose membrane by Mini-PROTEAN 3 Cell (Bio-Rad).

Membranes were incubated overnight at 4 °C with patient’s serum (IgE 1:20 dilution; IgG 1:35; IgG₄ 1:20). After three washes, the membranes were incubated with alkaline phosphatase-labelled monoclonal anti-human IgE (Sigma-Aldrich) at a 1:2000 dilution, IgG (Sigma-Aldrich) at a 1:2500 dilution and IgG₄ (Sigma-Aldrich) at a 1:1000 dilution. Finally, band detection was carried out with 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium (Sigma-Aldrich) for 20 min.

Histologically, the neoformation was composed by colic wall with marked edema of the submucosa and muscular layer with abundant transmural inflammatory infiltrate composed mainly by lymphocytes, plasma cells and over all eosinophils. The histological examination of the eosinophilic granuloma, of the overall size of about 0.15 cm, revealed the presence of a parasitic nematode, having a diameter of 0.48 × 0.30 mm, with a thin cuticle lacking of lateral alae. Polymiarian muscle cells, separated into four quadrants by the chords, having two wing-like distal lobes, were well visible; the nematode’s intestine was circular with a triangular lumen; columnar epithelial cells were disposed radially; no ventricular appendix and/or intestinal caecum was present. According to morphological features, the nematode was considered belonging to the genus Anisakis, but it was not possible to identify it at species level (Fig. 1c).

The mtDNA cox2 sequences obtained from the larval nematodes removed from the stomach in CC1 and the larva recovered in the intestine of CC2, showed 99 or 100% similarity with sequences of mtDNA cox-2 previously obtained for the species A. pegreffii [18] and retrievable from GenBank. These mtDNA cox2 sequences isolated from CC1 and CC2 were deposited in GenBank under the accession numbers: KY073403 e KY073404 (Fig. 2a).

In addition, according to the diagnostic positions found at the EF1 α − 1 region of the nDNA (409 bp), i.e. showing a T and a C in A. pegreffii, respectively, at position 186, and 286 the same larval specimens were assigned to the species A. pegreffii (Fig. 2b). Sequences of EF1 α − 1 region of nDNA were deposited in GenBank under the accession numbers: KY118808 and KY118809.

The extremely low amount of extracted DNA (<0,01 ng/μl, Qubit2.0 BR-dDNA kit), did not allow direct sequencing. On the other hand, the RT-PCR primers/probe system procedure, showed a consistent fluorescent signal at 510 nm (FAM dye), as defined for the species A. pegreffii [our unpublished data], but showing a relatively high C_t values (~34), as in Fig. 3, due to the low amount of DNA.

The RT-PCR assay was also used for the CC1 and CC2 DNA samples. The analysis was performed on CC1, CC2, CC3 in duplicates and, on human tissue (h), in triplicates. The limit of detection was established at a value under 0.0006 ng/μl. The reaction efficiency value was E = 2.013, with a slope value of σ = −3.29. No cross-reactivity with the genomic DNA from human blood sample was observed.

Among the three sera samples available in the present study, the serum from CC1 revealed a specific IgE and IgG₄ antibody reactivity to two antigens whose relative mobility (M_r) resulted around 37 kDa and 139 kDa, corresponding to Ani s 13-like and Ani s 7-like, respectively (Fig. 4). Conversely, the serum from CC2 showed IgG reactivity against Ani s 13-like and Ani s 7-like (Fig. 4). Finally, serum from CC3 showed both IgE and IgG reactivity against proteins/antigens whose M_r. was corresponding to Ani s 13-like, Ani s 7-like and with respect to another band recognized by both IgE and IgG, was observed, having a M_r. of 24 kDa, corresponding to Ani s 1-like (Fig. 4).

We documented one gastric and two intestinal cases of invasive anisakiasis caused by the species A. pegreffii. In Italy, previous cases of gastric [12, 22‐26] and intestinal [24, 27, 28] anisakiasis were reported. However, in those cases, despite the histological findings and morphological features allowed disclosing the presence of a larval Anisakis sp., the identification at the species level was not possible. By contrast, in the last decades, the application of molecular tools in this group of parasites has allowed the identification of larvae infecting humans, thus enlarging the knowledge about this fish-borne zoonosis [9‐16]. Furthermore, the implementation of the methodologies based on PCR, as those concerning DNA purification and extraction, have allowed to develop a method for the identification of A. pegreffii even from the paraffin-embedded tissues of surgically removed granulomas caused by the parasite [11]. This method was recently successfully applied also in an archival case in Croatia [13].

In the present study, a more sensitive method for the accurate and rapid identification of the etiological agent of human anisakiasis, based on RT-PCR hydrolisis probe, was utilized. This method detected up to 0.0006 ng/μl of DNA of the parasite. The RT-PCR probe assay was assessed for the first time, on a human tissue, (eosinophilic granuloma) whose etiological agent would have been undetectable as inferred by direct DNA sequencing, due to the low quantity of DNA available. This situation could also occur when bioptic tissues are collected by endoscopy from ulcers caused by the parasite in the gastric and/or intestinal mucosa, without direct observation of the parasite itself. Indeed, since the symptoms of anisakiasias are not pathognomonic, the acute form could be misdiagnosed. As a consequence, the infection could become chronic, leading to granuloma formation. Therefore, the early removal and identification of the worm makes the best prevention of formation of eosinophilic granuloma caused by the allergic reaction to the degenerated larva in the chronic infection.

The possibility to use sensitive and specific methods for the molecular diagnosis of human anisakiasis, such as a RT-PCR, as here proposed, will allow to enlarge the knowledge about its occurrence in human populations that regularly consume raw or undercooked fish.

In this respect, in Italy the consumption of home-made raw marinated anchovies that were previously not exposed to rapid freezing (−20 °C for 24 h), as compulsory by European Community regulation (EC no 1276/2011), represents the main risk to contract this fish-borne zoonosis [12]. Indeed, also in the cases, here described, the source of the infection was “home-made raw marinated anchovies”. On the other hand, anchovies Engraulis encrasicolus, is one the most important fish resource in the Mediterranean countries; this species was found to be heavily infected by Anisakis spp. larvae, even in its flesh (i.e. edible parts), in some fishing grounds of the Mediterranean Sea [29, 30]. In addition, it was demonstrated that a migration of A. pegreffii larvae from the viscera to the edible parts of the E. encrasicolus occurs post-mortem, with temperature- and time-dependent larval motility [29]. It is therefore important to increase the control of the temperature of the fish storage after landing, up to the domestic level.

A. pegreffii is not the only species able to cause human anisakiasis. In Japan, Umehara et al. [14] and Arai et al. [15] recognized A. simplex (s .s.) as the main etiological agent of anisakiasis in a large number of patients. The opposite situation was found in South Korea, where Lim et al. [16] described the infection in several patients as caused by A. pegreffii, but in one patient was due to A. simplex (s. s.), However, those findings could be related both to the fishing ground of the source of the infection (i. e. the infected fish consumed), and to the geographical distribution of the two sibling species of Anisakis. Indeed, while the Japanese waters are a sympatric area of the two species from South Pacific area, so far the species A. simplex (s. s.) has not been recorded in fish hosts; by contrast, in the same geographical area, A. pegreffii is present, also in co-infection with A. berlandi, in several fish and cetacean hosts [7, 8, 18].

Concerning the pathogenic potential of A. pegreffii in comparison with A. simplex (s. s.), according to the literature, A. pegreffii has been identified at molecular level in several cases of Gastro Allergic Anisakiasis (GAA) in Italy, characterized by the occurrence of urticaria manifestation and/or edema of oral mucosae [12]. Conversely, so far, no data have been recorded about the GAA due to A. simplex (s. s.), identified in clinical cases by molecular means [31]. Of course, this does not exclude the possibility that also A. simplex (s. s.) is able to elicit that immune response in humans causing GAA, also in consideration that some authors are reporting a higher capacity of this species to survive the pH condition of human stomach [32, 33] and to penetrate at higher percentage in the muscle of natural fish host [34, 35], but also in its accidental host (humans) [9‐16]. However, in the clinical cases here described, in spite of the fact that their etiological agent was A. pegreffii, we did not notice any allergic reaction; this finding is similar to that found by Lim et al. [16].

In the present study, we have also performed Immunoblotting assay to study the Anisakis-specific immune response in three different cases of anisakiasis due to A. pegreffii. Interestingly, the serum from the patient suffering of the eosinophilic granuloma (CC3) showed immunoreactivity at the Immunoblotting assay, at both IgE and IgG against the three most frequent antigens indicative of the infection by A. pegreffii (i.e. Ani s 1-like, Ani s 7-like and Ani s 13-like). Whereas, the sera from patient CC1 and CC2 did not react with antigen around Ani s 1-like. Hypotheses to explain these last findings could be related to: i) the serum available from the CC1 was taken just during patient’s hospitalization (i.e. few days after the infected fish intake); on the other hand, it has been reported that IgE reaction against Ani s 1 increased after 1 month from the infection [2, 36, 37]; ii) in both CC1 and CC2 cases, the parasite larva was found, at least, during the first 7 days after the infection.

Interestingly, the only gastric case of anisakiasis (i.e. CC1), here studied, showed IgG₄ reactivity against Ani s 7-like. Typical for a parasite-induced immunologic reaction, previous studies observed the concomitant production of both IgE and other immunoglobulin isotypes, such as IgG₄, which are mediated by the same Th-2 mechanism [38‐40]. Further, IgG₄ response to Ani s 7 has been suggested as a marker of gastric allergic anisakiasis due to A. simplex [40]. Thus, the IgG4 seropositivity against Ani s 7-like observed in patient CC1 seems to be in accordance with previous observation [40].

Finally, a band of immunoreactivity having a relative mobility (Mr) around 37 kDa, likely corresponding to Ani s 13-like [21], was observed in all the sera tested in the present study; the same antigen was found in all the sera from patients in a large survey carried out on IgE sensitization by A. pegreffii, in Italy [21]. This suggests that Ani s 13-like could be a good candidate in the diagnosis of human anisakiasis.