Table 1.
Primer sequences for HLECs qPCR.
Table 2.
Primer sequences for qPCR of lymphedema tissues.
Fig 1.
Gene expression of Mcoln1 in lymphedema tissue and normal tissue.
(A) Mcoln1 expression profiling by microarrays from GEO database (GSE4333). Three sets of microarrays with three replicates each for a total of 9 arrays were performed. The three conditions were normal tail skin (no intervention), lymphedema tail skin (due to surgical lymphatic vessel blockage), and surgical sham control tail skin (surgical incision with no lymphatic vessel blockage). (B) Data visualization was performed using Python to perform t-test. Mcoln1 expression is higher in the lymphedema group than in the normal group (p < 0.05).
Fig 2.
Pathological changes in the tail of lymphedema mouse model.
SHAM: Sham operation group; WT: Wild-type mouse model group; TRPML1−/−: TRPML1 gene-knockout mouse model group. (A) Comparison of actual tail conditions of lymphedema mice in each group on day 27. The tails of mice in the SHAM group showed no tissue edema, and the tails showed a natural shape. The tails of mice in the WT group showed unnatural bending and severe tissue edema. The tails of TRPML1−/− mice were able to naturally straighten, and the symptoms of tissue edema were significantly weaker than those of WT mice. (B) Diameter changes in the cross-section of mouse tail edema. From days 0 to 27, the tail diameters of mice in the SHAM group did not change, while those of mice in the WT and TRPML1−/− groups significantly increased on day 9, peaked on day 21, and then plateaued. The curve of the WT group rose (the difference between the WT group and SHAM group was significant, ***p < 0.001), while the curve of the TRPML1−/− group showed a more gradual increase, with the degree of whole tail edema being lower (the TRPML1−/− and WT groups were significantly different, **p < 0.01). (C) Hematoxylin and eosin (HE) staining results of the cross-section of lymphedema sites in mice. Images were viewed with an Olympus BX53 microscope with an Olympus UPlanFLN 4×/0.13 objective. The average interstitial thickness of the SHAM group was ~600 μm (black arrow). The average interstitial thickness in the WT group significantly increased to ~1300 μm (blue arrow). The average interstitial thickness of the TRPML1−/− group was smaller than that of the WT group, ~800 μm (green arrow). Scale bar = 200 μm. (D) HE staining results of the cross-sections of lymphedema sites in mice. Images were viewed with an Olympus BX53 microscope with an Olympus UPlanFLN 10×/0.30 objective. In the SHAM group, the cross-sectional structure of the tail was complete, compact, and regularly arranged, and there was almost no inflammatory cell infiltration. In the WT group, the staining structure was incomplete, the interstitial tissue was obviously expanded, the collagen fiber was seriously broken, and many inflammatory cells were infiltrated (red arrow). In the TRPML1−/− group, the main tissue structure was undamaged, and the degrees of tissue interstitial expansion and inflammatory cell infiltration were slight. Scale bar = 100 μm.
Fig 3.
Gene and protein expression of AQPs in HLECs.
(A) Gene expression of AQP1-12B in HLECs was assessed using qPCR with GAPDH as the housekeeping gene. AQP3 gene expression was the highest, followed by AQP5 expression. (B) Gene expression of AQP1-12B in HLECs was analyzed through agarose gel electrophoresis. The PCR products were approximately 100 bp, and the results were consistent with those of qPCR. (C) The effect of TRPML1 on the expression of AQP3, -5 protein was analyzed through western blot using GAPDH as a control. (D) Statistical analysis of western blot strips is shown. HLECs were treated for 24 h according to the following groups: NC: negative control; ML-SA1: 30 μM; ML-SI1: 25 μM; ML-SA1+ ML-SI1: 25 μM ML-SI1 and 30 μM ML-SA1 co-incubated; BafA1: 0.1 μM (E) Gene expression of AQP3, -5 in mouse lymphedema tissue was assessed using qPCR. Compared with that in the SHAM group, AQP3, -5 gene expression in lymphedema tissue of WT mice showed a tendency to increase.
Fig 4.
Localization changes in AQP3, -5 in HLECs and actin skeleton staining.
HLECs and transfected Cos-1 cells were treated in the following groups: NC: negative control; ML-SA1: 30 μM for 30 min; ML-SI1: 25 μM for 30 min; ML-SA1+ ML-SI1: 25 μM ML-SI1 and ML-SA1 co-incubated for 30 min; BafA1: 0.1 μM for 30 min. (A) Subcellular localization changes in AQP3 in HLECs. In the NC and BafA1 group, AQP3 fluorescence was mostly distributed in the cytoplasmic region. The AQP3 fluorescence of the ML-SA1 group was distributed diffusively throughout the cell and accumulated on the surface of the cell membrane. ML-SI1 alone or with ML-SA1 did not change the fluorescence range of AQP3. Scale bar = 10 μm. (B) Cos-1 cells were transfected with AQP3-GFP plasmids, and the membrane orientation of AQP3 under each treatment was observed. Only the ML-SA1-treated cell showed obvious high-density green fluorescent patches on the membrane surface. Scale bar = 10 μm. (C) Subcellular localization changes in AQP5 in HLECs. In the NC and BafA1 group, AQP5 fluorescence was slight and diffused around the cell. According to the fluorescence distribution of the ML-SA1 group, AQP5 accumulated to the cell membrane. ML-SI1 alone or with ML-SA1 did not change the fluorescence range of AQP5. Scale bar = 10 μm. (D) Membrane orientation changes in the overexpression of AQP5 in Cos-1 cells. Only the ML-SA1-treated cell showed obvious high-density green fluorescent patches on the membrane surface. Scale bar = 10 μm. (E) Phalloidin staining of the cytoskeleton in HLECs. In the NC group, the fluorescence of phallus cyclic peptide was bright, and the whole cytoskeleton was clear. In the ML-SA1 group, the fluorescence of phallus cyclic peptide was dim, and the cytoskeleton disappeared or began to break. ML-SI1 alone or with ML-SA1 did not cause changes in the cytoskeleton; BafA1 also did not show any significant effect on the cytoskeleton. Scale bar = 30 μm. (F) The membrane portions of AQP3 following activation of TRPML1 were measured by western blot. (G) The membrane portions of AQP5 following activation of TRPML1 were measured by western blot. HLEC membrane proteins and plasma proteins were extracted separately to analyze the distribution of AQP3, -5. After ML-SA1 stimulation, both AQP3, -5 were more detected in the components of membrane proteins, and AQP3, -5 were more present in the form of polymers on the cell membrane, while relatively less AQP3, -5 was detected in the cytoplasm.
Fig 5.
Fluorescence changes in HLECs stained by Cal-AM.
HLECs were treated in the following groups: NC: negative control; ML-SA1: 30 μM for 30 min; ML-SI1: 25 μM for 30 min; ML-SA1+ ML-SI1: 25 μM ML-SI1 and ML-SA1 co-incubated for 30 min; BafA1: 0.1 μM for 30 min. (A) The fluorescence of HLECs was recorded at the 26th second. Only the ML-SA1-treated cells showed a transient green fluorescence signal around the membrane after hypotonic stimulation (white arrow), while the other groups of cells did not show any change in fluorescence signal after hypotonic stimulation. Scale bar = 40 μm. (B) The fluorescence intensity of HLECs was detected using a microplate reader. Compared with that of the NC group, the fluorescence intensity of the ML-SA1 group was enhanced, and the difference was significant (**p < 0.01). There was no significant difference in fluorescence intensity in the other three groups compared with that in the NC group (p > 0.05).
Fig 6.
Effect of TRPML1 on inflammation in lymphedema mice.
SHAM: sham operation group; WT: wild type mice model group; TRPML1−/−: TRPML1 gene knockout mouse model group. (A) Representative figure of CD86+ macrophage infiltration in the cross-sections of lymphedema sites in mice. In the SHAM group, a few CD86+ macrophages infiltrated. In the WT group, CD86+ macrophages infiltrated (red arrow). In the TRPML1−/− group, the CD86+ macrophage infiltration was slight. Scale bar = 20 μm. (B) Statistical analysis of immunohistochemistry assay. The average of CD86+ macrophages in the WT group was higher than that in the SHAM group, while the average of macrophages in the TRPML1−/− group was significantly less than that in the WT group (*p < 0.05). (C, D) Serum IL-1β/IL-6 secretion was detected through ELISA. The concentration of IL-1β/IL-6 in the SHAM group was normal. Compared with that in the SHAM group, the IL-1β/IL-6 concentration in the WT group significantly increased. The concentration of IL-1β/IL-6 in the TRPML1−/− group was significantly lower than that in the WT group (*p < 0.05, ***p < 0.001). (E–J) The relative expression of IL-1β/IL-6/TGF-β1/col-1/vimentin/α-SMA genes in mouse lymphedema tissues was detected using qPCR. Compared with that of the SHAM group, the expression of IL-1β/IL-6/TGF-β1/col-1/vimentin/α-SMA genes in the WT group was significantly increased. Compared with that in the WT group, the expression of IL-1β/IL-6/TGF-β1/col-1/vimentin/α-SMA genes was significantly decreased in the TRPML1−/− group (*p < 0.05, **p < 0.01, ***p < 0.001).