Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Aug 28;158(5):1033-1044.
doi: 10.1016/j.cell.2014.06.048.

A Secreted Tyrosine Kinase Acts in the Extracellular Environment

Free PMC article

A Secreted Tyrosine Kinase Acts in the Extracellular Environment

Mattia R Bordoli et al. Cell. .
Free PMC article

Erratum in

  • Cell. 2014 Nov 6;159(4):955


Although tyrosine phosphorylation of extracellular proteins has been reported to occur extensively in vivo, no secreted protein tyrosine kinase has been identified. As a result, investigation of the potential role of extracellular tyrosine phosphorylation in physiological and pathological tissue regulation has not been possible. Here, we show that VLK, a putative protein kinase previously shown to be essential in embryonic development, is a secreted protein kinase, with preference for tyrosine, that phosphorylates a broad range of secreted and ER-resident substrate proteins. We find that VLK is rapidly and quantitatively secreted from platelets in response to stimuli and can tyrosine phosphorylate coreleased proteins utilizing endogenous as well as exogenous ATP sources. We propose that discovery of VLK activity provides an explanation for the extensive and conserved pattern of extracellular tyrosine phosphophorylation seen in vivo, and extends the importance of regulated tyrosine phosphorylation into the extracellular environment.


Figure 1
Figure 1. Conserved position of phosphorylated tyrosine in hemopexin domains reported in vivo in Phosphosite compendium
A) All sites shown are at orthologous positions within the hemopexin domain as defined by primary sequence conservation. B) Hemopexin domain structures for MMP1 MMDB ID: 30864 (Jozic et al., 2005) and MMP13 (MMDB ID: 57090)(Gomis-Ruth et al., 1996) and hemopexin (MMDB ID:56395) (Faber et al., 1995). Yellow arrow indicates reported site of tyrosine phosphorylation (Hornbeck et al., 2004; Raijmakers et al., 2010). See also Figure S1.
Figure 2
Figure 2. VLK is constitutively secreted and its glycosylation and secretory pathway localization are signal peptide dependent
A) Sequence of mouse VLK. Signal peptide is indicated in red, signal peptide cleavage site with an arrow. Sites of point mutations discussed in the text are marked in red with asterisk, the ProGly rich domain is underlined in dashed blue. The predicted kinase region is underlined in black. B) Detergent free lysates of 293T cells overexpressing GFP or VLK were treated with proteinase K (PK) alone or in combination with NP40; C) Detergent free lysates of HepG2 cells co-expressing VLK and VLK-SP were treated with proteinase K (PK) alone or in combination with NP40. D) VLK expression was determined in lysates and CM of 293T cells overexpressing VLK or VLK-SP. Tyrosine phosphorylation was detected following VLK immunoprecipitation. E) Cell lysates expressing either VLK or VLK-SP were treated with PNGase to remove N-linked glycosylations. F) Lysates of HepG2 cells were treated with PK alone or in combination with NP40, and endogenous VLK detected by Western blot. Cytoplasmic actin was measured as a marker of cytoplasmic protein. G) VLK expression and secretion was detected in lysates and CM of brefeldin A treated HepG2 cells. H) VLK expression was detected in lysates (Lys) and CM of HepG2 cells treated with brefeldin A, before and after ultracentrifugation. I) VLK protein levels were determined in CM of HepG2 cells stably transduced with either a scrambled shRNA (c) or two independent shRNAs targeting Vlk (#2, #5). Percentage of protein expression was quantified using ImageJ. See also Figure S2.
Figure 3
Figure 3. VLK is secreted from platelets and is responsible for tyrosine phosphorylation in the platelet releasate
A) Freshly isolated human platelets were treated with either 0.2 μM PMA or 0.5 μM TRAP for 15′. After centrifugation to remove cells, cell free releasate was incubated for an additional 15′ in the presence or absence of 2 mM ATP (Sup.+ATP). Black arrow indicates phospho-tyrosine band co-migrating with VLK. Asterisks mark bands that change in releasate with stimulation and ATP. B) Tyrosine phosphorylation is dependent on VLK and on endogenous ATP. Left Panel: Following platelet stimulation with TRAP for 4′, VLK was depleted from releasates by incubation with anti-VLK or control IgG and protein A sepharose for the indicated times, and then phosphorylation was stimulated by exogenous 2 mM ATP addition (+ATP) for 15′. Note that some VLK dependent phosphorylation occurs in the absence of exogenous ATP, presumably due to endogenous secreted ATP. Right panel: platelets were stimulated with TRAP in the presence of 100 Units of apyrase to degrade endogenous released ATP. After 4′ stimulation with TRAP, platelets were cleared by centrifugation and releasate incubated for additional time indicated, then analyzed by anti-pTyr Western blot. Asterisks indicate bands changing following treatments. C) Immunofluorescence of platelets with anti-VLK antibody shows puncate staining. D) Immunogold-EM staining with anti-VLK localizes VLK to alpha granules. Scale bar is 5 μm in C and 0.5 μm in D. See also Figure S3.
Figure 4
Figure 4. VLK phosphorylates secreted and transmembrane substrates
A) 293T cells were co-transfected with the indicated candidate substrate and either VLK wild-type or VLKKM. Tyrosine phosphorylation was detected following immunoprecipitation of candidate substrates from cell lysates. B) Tyrosine phosphorylation was analyzed in lysates of 293T cells co-expressing MMP1 and VLK, VLKKM, or VLK-SP following MMP1 immunoprecipitation. C) MMP13 was overexpressed in 293T cells stably transduced with either a scrambled shRNA or an shRNA targeting VLK (#5). Tyrosine phosphorylation was examined in CM following MMP13 immunoprecipitation. VLK expression was determined in cell lysates (Lys.) D) Freshly isolated human platelets were treated with 0.5 μM TRAP for 15′. After centrifugation to remove cells, protein expression and tyrosine phosphorylation were determined in pellets and supernatants. E) Detergent free lysates of 293T cells expressing the indicated VLK constructs were treated with PK alone or in combination with NP40. See also Figure S4.
Figure 5
Figure 5. VLK expression enhances cellular and secreted tyrosine phosphorylation
A) K4 synoviocytes stably over-expressing GFP, VLK or VLKKM were treated with 100 nM dasatinib for 6 hrs. Extent of tyrosine phosphorylation in lysates was assessed by immunoblotting. Asterisks indicate VLK-dependent phosphorylations. B) Tyrosine phosphorylation was detected in TCA-precipitated CM of K4 synoviocytes over-expressing VLK or VLKKM. See also Figure S5.
Figure 6
Figure 6. Baculovirus expressed VLK phosphorylates peptides and proteins
Wild type VLK, or VLK mutated at Lys66 (VLKKR) was purified from CM of a baculovirus expression system and tested for peptide or protein phosphorylation activity. A) Biotinylated peptides were incubated with purified VLK protein in the presence of 33[P]-γ-ATP, captured on streptavidin and counted by liquid scintillation counting. YPeptide: GRRYLQELQKEQ; SPeptide:GRRSLQELQKEQ, TPeptide GRRTLQELQKEQ; HpxPeptide: peptide derived from MMP1 hemopexin domain, QNVLHGYPKDI. B) Purified wild type or mutant recombinant ERP29 phosphorylated by purified recombinant wt VLK or VLK mutated at Lys166 (KR) analyzed by Laemmli gel electrophoresis and autoradiography. See also Figure S6.

Comment in

Similar articles

See all similar articles

Cited by 38 articles

See all "Cited by" articles

Publication types