Fig 1.
Nanobodies are specific for PRL-3 over the other PRL family members.
(A) Schematic of the process of generating and isolating anti-PRL-3 nanobodies (Created with BioRender.com). (B) Binding of each histidine-tagged nanobody (NB, at a concentration of 6.22 pmol) to 5 pmol PRL-1, PRL-2, or PRL-3 in 96-well plates. (C) The binding of each nanobody at the concentrations indicated to 5 pmol of each PRL was measured by indirect ELISA. Data are the absorbance at 450 nm after nanobody/PRL wells were washed and probed with His-HRP conjugated antibody. All assays were completed with two technical replicates and repeated in two biological replicates. Error bars represent standard deviation. ns = not significant, *p < 0.05, ***p < 0.001 by two-way ANOVA with Sidak’s multiple comparisons test. The number of amino acid changes compared to the most common anti-PRL-3 nanobody sequence is indicated by color coding.
Fig 2.
Nanobodies are specific to PRL-3 in immunofluorescence assays.
HCT116 colorectal cancer cells were transfected with CMV:GFP, CMV:GFP-PRL-1, CMV:GFP-PRL-2, or CMV:GFP-PRL-3 for 24 hours prior to cell fixation and permeabilization. Immunofluorescence assays were completed with 1:100 1 mg/mL nanobody 91 followed by 1:400 Alexa Fluor® 594-AffiniPure Goat Anti-Alpaca IgG, VHH domain, showing that nanobodies detect and co-localize with PRL-3 but not PRL-1 or PRL-2. We repeated this assay with nanobodies 4, 10, 16, 19, 26, and 84; all were similarly specific for PRL-3 over PRL-1 and PRL-2 (S2–S7 Figs).
Fig 3.
PRL-3 localization assessed by nanobody 26 and 19 is altered by N-terminal 3XFLAG and GFP tags.
(A) Immunofluorescence of HCT116 cells transfected with CMV:3XFLAG-PRL-3, CMV:GFP-PRL-3, CMV:HA-PRL-3, or CMV-PRL-3, as indicated. Cells were stained with anti-PRL-3 nanobody 26 or 19 (HA) followed by an anti-alpaca VHH coupled to Alexa594 secondary antibody for visualization. (B) ImageJ quantification of nanobody/PRL-3 staining. Groups were compared using a Mixed-effects analysis with Tukey’s Multiple Comparisons Test where **p < 0.01, ***p < 0.001, ****p<0.0001.
Fig 4.
Nanobodies selectively immunoprecipitate PRL-3 from HEK293T cell lysate and IP HA-PRL-3 comparatively to commercially available anti-PRL-3 antibodies.
(A) Nanobody 19 pulls down 3XFLAG-PRL-3 with minimal to no pulldown of 3XFLAG-PRL-1 or 3XFLAG-PRL-2. Successful nanobody coupling to Dynabeads was verified using an antibody against the C-terminal 6XHis-tag present on each nanobody. (B) Nanobody 19 pulls down HA-PRL-3 from HCT116 cells just as well as commercially available antibodies MAB3219 and sc-130355, and better than ab50276 and GTX100600. The input, transiently overexpressed HA-PRL-3, is shown separately. Nanobody 19 was coupled to superparamagnetic Dynabeads® M-270 Epoxy beads and used in immunoprecipitation assays with lysates from HEK293T cells transduced with 3XFLAG-PRL-1, -2 or -3 and HCT116 cells transiently transfected with HA-PRL-3. All anti-PRL-3 antibodies were incubated with HA-PRL-3 lysate and pulled down using Protein A and Protein G beads.
Table 1.
Kinetics of nanobody binding to PRL-3.
Fig 5.
Hydrogen Deuterium Exchange Mass Spectrometry (HDX-MS) defines nanobody 91 binding sites with PRL-3.
(A) PRL-3 in complex with nanobody 91 shows regions of increased (red) and decreased (blue) deuterium uptake, compared to apo-PRL-3. Heatmap indicates approximately 70% sequence coverage by mass spectrometry; gray areas represent portions of PRL-3 where data for deuterium exchange was not recovered. (B) Peptides 13–19 on PRL-3 were deprotected following nanobody binding, while peptides 56–79 and 132–146 showed decreases in deuterium uptake, reflecting more protection by nanobody 91 on PRL-3 in these regions.
Fig 6.
Anti-PRL-3 nanobodies partially interact with the PRL-3 active site and site of CNNM3 CBS-domain binding.
(A) Phosphatase activity of PRL-3 alone or in complex with nanobody 91, against a generic diFMUP substrate. DiFMUP fluoresces when phosphate is released. The graph shows normalized fluorescence to PRL-3 alone, n = 12 ± standard deviation. Assays were completed with six technical replicates and repeated in two biological replicates. Error bars represent standard deviation. ns = not significant, **p < 0.01, ***p<0.001 ****p < 0.0001 by one-way ANOVA with Dunnett’s multiple comparisons test. (B) PDB:5TSR where the CNNM CBS domain is colored in orange, and PRL-3 in blue. Specifically, the dark blue is the footprint for nanobody 91 binding (Fig 5), and cyan is the rest of the PRL-3 surface filling structure. (C) Controls for competition assays between CNNM3 CBS domain and nanobody 26. 3XFLAG-PRL-3 pulldown and co-immunoprecipitation controls include His-tagged nanobody 26 pulldown alone or HA-tagged CNNM CBS domain pulldown alone, to show both proteins Co-IP with PRL-3. (D) Immunoprecipitation competition assays 1–3. 3XFLAG-tagged PRL-3 was complexed with anti-FLAG beads and either the HA-tagged CBS domain of CNNM3 (1), histidine-tagged nanobody 26 (2), or neither for 1 hour. After 1 hr incubation, nanobody 26-His (1) CBS-HA (2), or both proteins (3) were added to the complex for the second hour. L, ladder; I, input. Antibodies used for western blot are shown. Quantification is of CBS-HA and nanobody 26-His pulldown normalized to 3XFLAG-PRL-3 immunoprecipitation lane by ImageLab normalization analysis.