The Enzymatic Activities of Nucleoside Triphosphate Diphosphohydrolases Towards Tenofovir Diphosphate and Their Spatial Distributions in Human Colorectal Tissue

Abstract

Tenofovir (TFV; prescribed as the tenofovir disoproxil fumarate prodrug) must be phosphorylated to form a competitive inhibitor of HIV, TFV-diphosphate (TFV-DP). Previously, we reported heterogeneity in TFV-DP distribution in colorectal tissue (a putative site of HIV infection) sections collected from human participants receiving a TFV-containing enema. This observed heterogeneity is likely multifactorial. TFV-DP is structurally similar to ATP. It is known that nucleotidases such as nucleoside triphosphate diphosphohydrolases (NTPDases) dephosphorylate ATP. Therefore, we hypothesized that NTPDase-mediated dephosphorylation plays a role in regulating TFV-DP levels in colorectal tissue. To test this, we incubated recombinant NTPDase proteins (NTPDase 1, 3, 4, 5, and 8), individually, with TFV-DP to determine the abilities of these enzymes to dephosphorylate TFV-DP. Following incubation, TFV-DP levels were measured using ultra-high performance liquid chromatography tandem mass spectrometry. We observed TFV-DP dephosphorylation in the presence of NTPDase 1 or 3 enzymes. Further, a malachite green phosphate assay that measures the release of inorganic phosphate was employed to further confirm the activities of NTPDase 1 and 3 towards TFV-DP. Immunoblotting revealed that NTPDase 1 and 3 proteins are indeed expressed in human colorectal tissue lysates. To visualize the distributions of NTPDases across human colorectal tissue, we employed Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Imaging (MALDI MSI). Initially, trypsin digestions on a range of recombinant NTPDases including 1 and 3 were carried out to detect and identify their tryptic peptides that could be targeted for detection in MALDI MSI. From these digestions, NTPDase 1 yielded peptides, SQHQETPVYLGATAGMR, SLSNYPFDFQGAR, and IITGQEEGAYGWITINYLLGK whereas NTPDase 3 exhibited the following peptides: YGIVLDAGSSR, GSGISSYGNNPQDVPR, and MFTVLTR. Five tryptic peptides corresponding to NTPDase 5 (which did not exhibit enzymatic activity towards TFV-DP) were identified: LYTHSYLGFGLK, LHQPEEVQR, GSFYAFSYYYDR, QGAETVQGLLEVAK, and LATLGALETEGTDGHTFR. Next, the abovementioned MALDI MS methodology was applied to directly visualize tryptic peptides of NTPDase 1, 3 and 5 in human colorectal tissue sections. In these studies, human colorectal biopsies were collected and sectioned to 20-micron thickness. Tissue sections were subjected to serial washing steps using ethanol. Trypsin and matrix (α-cyano-4-hydroxycinnamic acid) were deposited onto the tissue using a robotic sprayer. We detected tryptic peptides, SQHQETPVYLGATAGMR, and LATLGALETEGTDGHTFR corresponding to NTPDase 1 and 5, respectively. The distribution profiles of NTPDase 1 and 5 were generated using the m/z values of above peptides. The identities of the detected peptides were validated using mass spectrometry-based proteomics analyses. Thus, in this work, we established a MALDI MSI methodology for simultaneous visualization of tryptic peptides corresponding to NTPDases 1 and 5. Results from our in vitro studies suggest that NTPDase 1 and 3 may contribute to the regulation of TFV-DP levels.