Characterization and PCR Detection Of Binary, Pir-Like Toxins from Vibrio parahaemolyticus Isolates that Cause Acute Hepatopancreatic Necrosis Disease (AHPND) in Shrimp

Abstract

Unique isolates of Vibrio parahaemolyticus (VPAHPND) have previously been identified as the causative agent of acute hepatopancreatic necrosis disease (AHPND) in shrimp. AHPND is characterized by massive sloughing of tubule epithelial cells of the hepatopancreas (HP), proposed to be induced by soluble toxins released from VPAHPND that colonize the shrimp stomach. Since these toxins (produced in broth culture) have been reported to cause AHPND pathology in reverse gavage bioassays with shrimp, we used ammonium sulfate precipitation to prepare protein fractions from broth cultures of VPAHPND isolates for screening by reverse gavage assays. The dialyzed 60% ammonium sulfate fraction caused high mortality within 24-48 hours post-administration, and histological analysis of the moribund shrimp showed typical massive sloughing of hepatopancreatic tubule epithelial cells characteristic of AHPND. Analysis of the active fraction by SDS-PAGE revealed two major bands at marker levels of approximately 16 kDa (ToxA) and 50 kDa (ToxB). Mass spectrometry analysis followed by MASCOT analysis revealed that both proteins had similarity to hypothetical proteins of V. parahaemolyticus M0605 (contig034 GenBank accession no. JALL01000066.1) and similarity to known binary insecticidal toxins called 'Photorhabdus insect related' proteins A and B (Pir-A and Pir-B), respectively, produced by the symbiotic, nematode bacterium Photorhabdus luminescens. In in vivo tests, it was shown that recombinant ToxA and ToxB were both required in a dose dependent manner to cause AHPND pathology, indicating further similarity to Pir-A and -B. A single-step PCR method was designed for detection of the ToxA gene and was validated using 104 bacterial isolates consisting of 51 VPAHPND isolates, 34 non-AHPND VP isolates and 19 other isolates of bacteria commonly found in shrimp ponds (including other species of Vibrio and Photobacterium). The results showed 100% specificity and sensitivity for detection of VPAHPND isolates in the test set.

Fig 1. Examples of histopathological sections of hepatopancreatic (HP) tissue from moribund shrimp treated by reverse gavage with 60% AS fractions from V. parahaemolyticus isolates.

(A) A longitudinal section of HP tissue from pre-challenged shrimp showing normal histology with the lumens enclosed by epithelial cell layers comprised of non-vacuolated deeply basophilic (purple stained), embryonic cells (E-cells) at the distal end of the tubule that progress in the proximal direction into a mixture of B-cells with large, single vacuoles, R-cells with multiple vacuoles and F-cells that are non-vacuolated and deeply basophilic. (B) A section of normal HP tissue from the PBS negative control shrimp showing normal tubules mostly in longitudinal section except for a few tubules at the outer (distal) portion of the HP where they are cut in cross-section. The tubule lumens are surrounded by epithelial cells similar to those in (A). (C) Tangential section of HP tissue from shrimp treated with non-AHPND S02 preparation and showing normal HP and showing the same cell types as in (A) and (B). (D) Section HP tubules (mostly in cross-section) from shrimp treated with 5HP preparation and showing AHPND pathology characterized by absence of normal epithelia containing B-cells, R-cells and F-cells as seen in (A) to (C) and instead by massive sloughing of epithelial cells into tubule lumens in the absence of bacteria. The inset shows a magnification of the sloughed epithelial cells in a tubule lumen. (E) Section of HP tubules (cross-section) from shrimp treated with CN preparation and showing AHPND pathology similar to that in (D) but more severe in that all of the tubule lumens are completely filled with sloughed cells except for two tubules cut in cross-section through the E-cell region.

Fig 2. SDS-PAGE analysis of 60% AS fractions from broth of non-AHPND isolate S02 (Lane 2), and VP_AHPND isolates 5HP (Lane 3) and CN (Lane 4).

Lane 1: 60% AS from culture broth without bacteria. There are 2 major bands (ToxA at ~16kDa and ToxB at ~50 kDa) present in lanes for the VP_AHPND isolates 5HP and CN, but absent or at low level in non-AHPND isolate S02.

Fig 3. Bacterial expression of ToxA and ToxB.

(A) ToxA expressed with a 6-His tag and purified by Ni-NTA affinity chromatography. Lane 1: Bacterial cell lysate from a non-induced bacterial culture; Lane 2: Bacterial cell lysate from an IPTG-induced culture; Lane 3: Eluted protein from the Ni-NTA column. The deduced molecular weight for ToxA-His was 12.7 kDa. (B) ToxB was expressed as a GST-fusion protein. Lane 1: Bacterial cell lysate from a non-induced culture; Lane 2: Bacterial cell lysate from an IPTG-induced culture; Lane 3: Eluted fraction from Sepharose 4B beads; Lanes 4&5: Fraction eluted from Sepahrose 4B after thrombin-cut. The estimated molecular weights for GST-ToxB and ToxB were approximately 76 and 50 kDa, respectively.

Fig 4. The effect of recombinant ToxA and ToxB administration on shrimp.

(A) Graph of cumulative mortality up to 48 h from single and mixed toxins at various concentrations. The graph contains results from two experiments, each with a BSA control (10 ug/g shrimp) (SD = 0) and with co-administration of 10 ug/g each of ToxA+ToxB (SD bars). The other treatments were not duplicated (no SD bars). (B) Graph of mortality at 24 h post administration versus mixed toxin concentration, yielding a linear regression line and rough LD₅₀ for the mixed toxins of approximately 6 ug/g of each.

Fig 5. Examples of HP histology of moribund shrimp administered single or combined doses of ToxA and ToxB by reverse gavage.

(A) Tissue sections from a BSA (10/20 μg/g) negative control shrimp (tubule longitudinal sections) and (B) a ToxA (5 μg/g) treated shrimp (tubule cross sections) showing only normal histology with morphology and cell types as described in Fig 1A to 1C. (C) ToxB only (5 μg/g) and (D) ToxA+ToxB (2 μg/g each) showing mostly normal histology, but with some thin HP tubule epithelia (black arrows) when compared to epithelia of normal thickness (grey arrows). (E) ToxA+ToxB (5 g/g each) showing enlarged HP tubules (compared to A-C) with collapsed epithelia (black arrows) but no cell sloughing. (F) ToxA+ToxB (10 μg/g each) showing massive sloughing (black arrows) and dissolution of HP tubule epithelial cells (i.e., severe AHPND histopathology).

Fig 6. Agarose gel of PCR amplicons from VP_AHPND using the AP3 method.

Lane M; DNA marker, Lane N: negative control; Lanes 1–3: Positive amplicons (333 bp) from 3 isolates of VP_AHPND bacteria; Lanes 4–10: No amplicons from 8 non-AHPND bacteria; Lane P: positive control (333 bp).