(A) Schematic diagram of the LYVE-1 locus, the targeting construct and the mutant allele. Homologous recombination of the targeting vector introduces the neo gene downstream of the LYVE-1 promoter, deleting exons 2 and 3 of the LYVE-1 gene. Green boxes indicate the coding exons, whereas the dashed lines indicate the regions of identity between the LYVE-1 locus and the targeting vector. tk, selectable thymidine kinase gene; WT, wild-type. S, SspI. (B) Southern blot analysis of SspI-digested genomic DNA from targeted ES cells using an 856-bp 5’ external probe generated from PCR primers UF and UR. A 6.2-kb restriction fragment is diagnostic for the mutated allele, whereas a 4.2-kb fragment is generated from the wild-type allele. +/+, wild-type; +/−, heterozygous; −/−, homozygous mutant. (C) Genotyping by PCR analysis of genomic tail DNA. Primer pairs CF1/CR2 and CF1/neo-PR were used to detect the wild-type and mutant alleles, respectively. (D) Western blot analysis of proteins (20 µg) from lung tissue using LYVE-1 antibody detects a 60-kD band in wild-type and heterozygous samples, but not in homozygous mutant samples.

FITC-dextran of 2 million Da was injected subcutaneously into the tail (A) and ear (B) of wild-type and mutant mice. Vessels were imaged using fluorescent microscopy. (A) The measured parameters are vessel diameter (d), and horizontal (h) and vertical (v) mesh sizes. (B) Arrows indicate fluorescent lymphatic vessels in the mouse ear, whereas arrowheads indicate blood vessels. Results shown are from 1 mouse in each group representative of 5 mice per group. WT = wild-type; LY KO = LYVE-1^−/−; CD KO = CD44^−/−; and dKO = LYVE-1^−/−/CD44^−/−. Images are 1.75 mm wide.

(A) Immunostaining of lymphatic vessels (arrows) with anti-VEGFR-3 monoclonal antibodies in kidney tissue show no differences in the four animal types. (B) Lymphatic vessels (arrows) in liver tissue appear similar in wild-type and mutant mice using immunostaining with anti-podoplanin antibodies. WT = wild-type; LY KO = LYVE-1^−/−; CD KO = CD44^−/−; and dKO = LYVE-1^−/−/CD44^−/−. Scale bar = 50 µm.

A) Lymph flow velocity in the tail lymphatic vessel network in wild-type (n = 8), LYVE-1 KO (n =7) mice, CD44 KO (n = 8) and LYVE-1/CD44 double KO (n = 8) mice was determined by residence time distribution analysis and showed no difference amongst the four groups (p > 0.3). Data are presented as mean ± standard error. B) Carrageenan-induced paw edema is significantly enhanced in LYVE-1/CD44 double knockout mice. λ-carrageenan (30 µl) was injected into the right paw of mice and edema was determined by paw thickness during daily monitoring. WT (n = 20), LY KO (n =11) mice, CD44 KO (n = 7) and dKO (n = 9) mice. *p < 0.05 dKO compared to WT; # p < 0.05 dKO compared to LYVE-1 KO. (C) Paws were removed 4 days after λ-carrageenan injection, freeze-dried and hyaluronic acid levels were determined using an HA test kit. HA levels are expressed as µg HA protein per mg of dry weight tissue and showed no differences amongst the groups (p < 0.05). WT (n = 6), LY KO (n = 8), CD44 KO (n = 6) and dKO (n = 6). Data are presented as the mean ± standard error. D) Mice were given an intraperitoneal injection of thioglycollate and subjected to peritoneal lavage 4 days later. Total numbers of viable cells were counted by using trypan blue dye exclusion method. WT (n = 6), CD44 KO (n = 4), LY KO (n =7) and dKO (n = 3) mice. E) Cells were subjected to differential staining (Giemsa) and macrophages, polymorphic neutrophils (PMNs) and lymphocytes were counted. Data are presented as the mean ± standard error.