Quickchange Lightning Site Directed Mutagenesis as Written in a Peer-review

• Journal Listing
• Commun Biol
• v.4; 2021
• PMC8055892

Commun Biol. 2021; 4: 482.

A collagen glucosyltransferase drives lung adenocarcinoma progression in mice

Hou-Fu Guo,^one Neus Bota-Rabassedas,^one Masahiko Terajima,² B. Leticia Rodriguez,¹ Don L. Gibbons,¹ Yulong Chen,¹ Priyam Banerjee,¹ Chi-Lin Tsai,³ Xiaochao Tan,¹ Xin Liu,¹ Jiang Yu,¹ Michal Tokmina-Roszyk,^iv Roma Stawikowska,⁴ Gregg B. Fields,⁴ Mitchell D. Miller,⁵ Xiaoyan Wang,³ Juhoon Lee,^{6, seven} Kevin N. Dalby,^{6, 7} Chad J. Creighton,^{eight, nine} George N. Phillips, Jr,^{5, 10} John A. Tainer,³ Mitsuo Yamauchi,² and Jonathan Chiliad. Kurie ^one

Hou-Fu Guo

¹Department of Thoracic/Head and Neck Medical Oncology, The Academy of Texas Physician Anderson Cancer Eye, Houston, TX USA

Neus Bota-Rabassedas

¹Department of Thoracic/Head and Neck Medical Oncology, The University of Texas Medico Anderson Cancer Center, Houston, TX USA

Masahiko Terajima

²Sectionalisation of Oral and Craniofacial Health Sciences, Adams Schoolhouse of Dentistry, University of North Carolina at Chapel Loma, Chapel Loma, NC United states of america

B. Leticia Rodriguez

¹Section of Thoracic/Caput and Neck Medical Oncology, The University of Texas Md Anderson Cancer Center, Houston, TX U.s.a.

Don L. Gibbons

¹Department of Thoracic/Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX U.s.a.

Yulong Chen

¹Section of Thoracic/Head and Neck Medical Oncology, The Academy of Texas Md Anderson Cancer Centre, Houston, TX USA

Priyam Banerjee

¹Section of Thoracic/Head and Neck Medical Oncology, The University of Texas Dr. Anderson Cancer Eye, Houston, TX USA

Chi-Lin Tsai

^threeDepartment of Molecular and Cellular Oncology, The Academy of Texas Physician Anderson Cancer Center, Houston, TX USA

Xiaochao Tan

^aneSection of Thoracic/Head and Neck Medical Oncology, The Academy of Texas Dr. Anderson Cancer Eye, Houston, TX The states

Xin Liu

¹Department of Thoracic/Head and Neck Medical Oncology, The Academy of Texas MD Anderson Cancer Center, Houston, TX USA

Jiang Yu

^aneSection of Thoracic/Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX United states

Michal Tokmina-Roszyk

^ivInstitute for Human being Health & Affliction Intervention (I-Wellness) and Department of Chemistry & Biochemistry, Florida Atlantic University, Jupiter, FL U.s.a.

Roma Stawikowska

^fourPlant for Human Health & Affliction Intervention (I-Health) and Section of Chemistry & Biochemistry, Florida Atlantic University, Jupiter, FL USA

Gregg B. Fields

^fourInstitute for Human Health & Disease Intervention (I-HEALTH) and Department of Chemistry & Biochemistry, Florida Atlantic University, Jupiter, FL U.s.a.

Mitchell D. Miller

⁵Department of Biosciences, Rice University, Houston, TX USA

Xiaoyan Wang

³Department of Molecular and Cellular Oncology, The University of Texas Medico Anderson Cancer Center, Houston, TX United states

Juhoon Lee

^half-dozenSectionalisation of Medicinal Chemistry, Targeted Therapeutic Drug Discovery and Development Program, College of Pharmacy, The University of Texas at Austin, Austin, TX USA

^viiDivision of Chemic Biology & Medicinal Chemistry, College of Pharmacy, The Academy of Texas at Austin, Austin, TX The states

Kevin N. Dalby

⁶Sectionalisation of Medicinal Chemistry, Targeted Therapeutic Drug Discovery and Development Program, College of Pharmacy, The Academy of Texas at Austin, Austin, TX U.s.a.

⁷Sectionalisation of Chemical Biology & Medicinal Chemistry, College of Pharmacy, The University of Texas at Austin, Austin, TX Us

Chad J. Creighton

⁸Section of Medicine, Dan L. Duncan Cancer Center, Baylor Higher of Medicine, Houston, TX USA

⁹Department of Bioinformatics and Computational Biological science, The University of Texas Dr. Anderson Cancer Center, Houston, TX Usa

George Due north. Phillips, Jr

^fiveSection of Biosciences, Rice University, Houston, TX Us

¹⁰Department of Chemistry, Rice University, Houston, TX U.s.a.

John A. Tainer

³Section of Molecular and Cellular Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX United states of america

Mitsuo Yamauchi

ⁱⁱDivision of Oral and Craniofacial Health Sciences, Adams Schoolhouse of Dentistry, University of North Carolina at Chapel Hill, Chapel Hill, NC The states

Jonathan M. Kurie

^oneDepartment of Thoracic/Caput and Neck Medical Oncology, The Academy of Texas Medico Anderson Cancer Center, Houston, TX United states

Received 2022 Jun 5; Accustomed 2022 Mar 8.

Supplementary Materials

Peer Review File

GUID: 8E9AE0D6-18AB-4F6D-9966-9E80FF6E6A19

Supplementary Information

GUID: 0B4CDBE8-8B66-45BD-8279-7029DF03970D

Description of Boosted Supplementary Files

GUID: 9B4666D4-610D-4022-ACC1-89F9A3046808

Supplementary Data 1

GUID: 4469ED49-D3EC-40A4-9513-1AB43D36C44C

Reporting Summary

GUID: D1EFCC20-B858-4390-8B6D-96E4BA9BBDC9

Data Availability Argument

Source data underlying plots shown in figures are provided in Supplementary Data 1. All other data, if any, will be available upon reasonable asking.

Abstract

Cancer cells are a major source of enzymes that modify collagen to create a stiff, fibrotic tumor stroma. Loftier collagen lysyl hydroxylase 2 (LH2) expression promotes metastasis and is correlated with shorter survival in lung adenocarcinoma (LUAD) and other tumor types. LH2 hydroxylates lysine (Lys) residues on fibrillar collagen's amino- and carboxy-terminal telopeptides to create stable collagen cross-links. Here, we show that electrostatic interactions between the LH domain active site and collagen make up one's mind the unique telopeptidyl lysyl hydroxylase (tLH) activeness of LH2. However, CRISPR/Cas-9-mediated inactivation of tLH activity does not fully recapitulate the inhibitory effect of LH2 knock out on LUAD growth and metastasis in mice, suggesting that LH2 drives LUAD progression, in part, through a tLH-contained mechanism. Protein homology modeling and biochemical studies identify an LH2 isoform (LH2b) that has previously undetected collagen galactosylhydroxylysyl glucosyltransferase (GGT) activity determined by a loop that enhances UDP-glucose-binding in the GLT active site and is encoded by alternatively spliced exon 13 A. CRISPR/Cas-9-mediated deletion of exon 13 A sharply reduces the growth and metastasis of LH2b-expressing LUADs in mice. These findings identify a previously unrecognized collagen GGT activity that drives LUAD progression.

Subject terms: Enzyme mechanisms, Cancer microenvironment, Molecular modelling, Glycobiology, Metastasis

Introduction

Fibrillar collagens play a cardinal role in maintaining tissue integrity and function¹. Their structural, biochemical, and mechanical properties are regulated, in part, by post-translational modifications of Lys residues². Fibrillar collagens have a central triple-helical construction ("helical domain") and amino- and carboxy-terminal not-helical "telopeptide" domains. Both Helical Lys (hLys) and Telopeptidyl Lys (tLys) residues tin exist hydroxylated to course helical hydroxylysine (hHyl) and telopeptidyl hydroxylysine (tHyl). tLys and tHyl are oxidized into allysine and hydroxyallysine, respectively, by lysyl oxidases. These aldehydes (allysine and hydroxyallysine) then undergo a series of condensation reactions with allysine, Lys, Hyl, and histidine (His) residues to course collagen cross-links that stabilize collagen fibrils and matrices. Hydroxylation of tLys into tHyl has lilliputian touch on on cross-link density but, upon oxidation, leads to the formation of a group of structurally singled-out stable collagen cross-links called Hyl aldehyde-derived collagen cross-links (HLCCs) that generate tensile strength in load-bearing skeletal tissues like bone and cartilage². Following oxidation and condensation, unhydroxylated tLys forms Lys aldehyde-derived collagen cross-links (LCCs) normally seen in soft tissues. hHyl residues in collagen tin can exist further modified by a 2-step glycosylation (galactosylation and glucosylation) to regulate how collagens interact with collagen receptors on cells^4–six. Reduced Lys hydroxylation and/or glycosylation underlie inherited connective tissue disorders^vii–11, and aberrantly high HLCC product contributes to fibrotic diseases and cancer metastasis^{three,12–15}.

Key collagen modifications are governed by lysyl hydroxylase family members (LH1-3) that have a conserved dual enzymatic domain architecture but are functionally distinct¹⁵. While all three LH family unit members hydroxylate helical Lys (hLys) residues in x-Lys-glycine sequences, LH3 is reportedly unique in its ability to role in tandem with GLT25D1/ii to catechumen helical Hyl residues into 1,two-glucosylgalactosyl-5-hyl through two consecutive reactions: GLT25D1/ii-mediated O-linked conjugation of galactose, and LH3-mediated conjugation of glucose to galactosyl-5-Hyl^16–19. In add-on, LH2 is unique in its ability to hydroxylate tLys residues, leading to the germination of stable HLCCs, as demonstrated by evidence that inactivating mutations of the LH2-encoding factor PLOD2 in Bruck Syndrome are associated with HLCC deficiencies and astringent skeletal abnormalities^7,9. In this written report, we sought to elucidate the structural basis for LH2's telopeptidyl LH (tLH) activity and to decide whether tLH activity underlies the pro-metastatic activeness of LH2 in LUAD.

Results

LH2-dependent LUAD models

We implemented an immunocompetent LUAD model^xx to place structural features of LH2 that promote tumor growth and metastasis. Relative to parental K-ras/Tp53-mutant 344SQ LUAD cells, 344SQ cells subjected to CRISPR/Cas-ix-mediated Plod2 knockout (KO) generated orthotopic LUADs that were smaller and less metastatic to mediastinal lymph nodes and the contralateral lung (Fig.1a-c, Supplementary Fig. one). Compared to parental cells, Plod2 KO 344SQ cells demonstrated no loss of proliferative activity in monolayer culture (Fig.1d) but generated multicellular aggregates that were less invasive in collagen gels (Fig.1e, f), which is in line with show that LH2 promotes LUAD invasive action³.

LH2 is critical for tumor invasion and metastasis.

a LH2 mRNA levels were measured by quantitative real-time polymerase chain reaction (qPCR) (bar graph, due north = iii). LH2 poly peptide levels were determined by western blot analysis (gels). Parental (WT) and CRISPR (clustered regularly interspaced short palindromic repeats)/Cas-nine (CRISPR associated protein 9)-mediated Plod2 knockout (KO) 344SQ cells. Actin used as loading control. b Cartoon of metastatic spread in immunocompetent mice bearing 344SQ orthotopic lung tumors. Mice were injected intra-thoracically with parental (WT) or Plod2 knockout (KO) 344SQ cells to generate a unmarried orthotopic lung tumor. After 2 weeks, mice were necropsied and visible tumors on the pleural surface of the contralateral lung were counted. c Numbers of visible main lung tumors (left dot plot, north = 7) and metastases to mediastinal nodes (eye plot, north = 7) or contralateral lung (right plot, north = 7) per mouse (dot). 344SQ cells described in (a). d Relative cell densities in monolayer culture adamant at each time point past WST-1 assays (due north = half-dozen, 7 or eight). e Cartoon of multicellular aggregates generated in microwell plates, transferred to collagen gels, and examined for formation of invasive projections. f Fluorescence micrographs of multicellular aggregates containing RFP-tagged 344SQ cells (images). Invasive projections (arrows). Left plot, number of invasion projections per aggregate (dot, n = 11 or 19). Correct plot, length of each invasive projection (dot, n = 110 or 191). KO aggregates demonstrated reduced invasive project length. g and h The percentages of T cells (g) and myeloid cells (h) in subcutaneous 344SQ tumors were measured past flow cytometric analysis. CD8 + T cells (CD8 + , n = 6 or 7), exhausted CD8 + T cells (PD-1 + TIM3 + , n = vi), regulatory T cells (FoxP3 + CD25 + , north = 6), myeloid-derived suppressor cells (CD11b + Gr-1 + , n = 6 or 7), dendritic cells (CD11c + F4_80- Gr-ane-, n = 6 or 7). Results are hateful values (±S.D.) from replicate samples. Error bars indicate ±Southward.D. p values, 2-tailed Educatee's t examination.

Enhanced intra-tumoral fibrosis is associated with reduced T jail cell infiltration, M2 macrophage polarization, and increased recruitment of regulatory T cells and myeloid-derived suppressor cells that inhibit CD8⁺ T cell immunity²¹. Allowed cells express collagen receptors that accept immunosuppressive functions²². To determine whether LH2 influences intra-tumoral immunity, we quantified immune jail cell subsets in subcutaneous tumors generated by Plod2 KO or parental 344SQ cells and identified alterations in T cell and myeloid cell subsets that were consistent with an anti-tumor response in Plod2 KO tumors (Fig.1g, h). Thus, LH2 influences collagen's immunosuppressive functions.

Electrostatic interactions specify tLH activity

Given that the LH domain active site is highly conserved across LH family members²³, we reasoned that residues extrinsic to LH2'southward active site decide tLH activity. To identify the domain in which those residues reside, nosotros performed rescue experiments on LH2-deficient MC3T3-E1 (MC) osteoblasts and found that residues required for HLCC reconstitution reside in LH2's LH domain (Fig.2a-1000, Supplementary Fig. 2). LH2 homology modeling based on a mimiviral tLH domain crystal construction (Fig.2h) identified two basic amino acid residues (R680 and R682) adjacent to the LH2 active site that are dramatically different in charge and hydrophobicity from the corresponding amino acids in LH1 and LH3 (Fig.2h, i). Furthermore, highly acidic aspartate and glutamate residues are positioned adjacent to tLys residues on fibrillar collagens (Supplementary Fig. 3). Replacing R680 and R682 on LH2 with the corresponding residues on LH1 (LH2-EP) ablated tLH action (Fig.2j), whereas activeness on type I collagen, which contains hLys residues, was relatively preserved (Fig.2k). Conversely, replacing collagen telopeptide'southward two acidic residues with alanine reduced the activity of wild-type, but non LH1-mimic, LH2 (Fig.2l, yard). These information suggest that LH2'due south unique tLH activeness is determined past electrostatic interactions between LH2 and collagen telopeptides.

LH2 t-LH specificity is adamant by electrostatic interactions with collagens.

a Pattern of LH constructs. Glycosyltransferase (GLT), accessory (AC), and lysyl hydroxylase (LH) domains. LH2'due south LH domain replaced with that of LH1 (LH2/one). b LH2 poly peptide levels were determined by western absorb assay of MC-3T3 (MC) cells that were stably co-transfected with LH2 shRNA (shLH2) and empty vector (-) or vectors expressing LH2, LH1, or LH2/1. Actin used as loading control. c–g Collagen cross-link quantification of matrices derived from MC-3T3 (MC) cells described in (b). Dihydroxylysinonorleucine (DHLNL, c), pyridinoline (Pyr, d), histidinohydroxymerodesmosine (HHMD, e), hydroxylysinonorleucine (HLNL, f), and the HLCC-to-LCC ratio (grand). The HLCC-to-LCC ratio was calculated every bit (DHLNL + Pyr)/HHMD. h LH2b LH domain structure was modeled using a homology-modelling server SWISS-MODEL. The construction model identifies a cluster of arginine residues (marine) most the LH2 active site. Fe²⁺ molecule (orange brawl) in active site. i Amino acid sequence alignment of human LHs. R680 and R682 in LH2 (pointer heads) are non in LH1 and LH3. j–m LH activity of LH2 and LH2-EP on a trimeric telopeptide (j, l, m) or type I collagen (k). Acidic residues at i-1 and i-2 positions in telopeptide (Telo) were mutated to alanine (TeloAA). LH activity was measured by detecting succinate production with an adenosine triphosphate (ATP)-based luciferase analysis. n and o Numbers of visible primary lung tumors (left plot) and metastases to mediastinal nodes (eye plot) or contralateral lung (right plot). Mice were injected intra-thoracically with parental (WT) or CRISPR/Cas-9-edited 344SQ cells. Plod2 mutations ablate LH2'due south tLH activity owing to loss of dimerization (L735D) (due north) or Ironⁱⁱ⁺-binding (D689A) (o). Results are hateful values (±Due south.D.) from replicate samples. For (c–g) and (j–chiliad), n = three. For (n and o), due north = ix or 10. Mistake bars bespeak ±S.D. p values, 2-tailed Student'south t examination.

tLH inactivation does not recapitulate the effect of Plod2 KO

Based on the temporal relationship betwixt stable collagen cross-link accumulations and enhanced tumor cell invasion and metastasis^{three,13,24–26}, we postulated that tLH activity accelerates LUAD progression. To test this hypothesis, nosotros ablated tLH activity in 344SQ cells past introducing mutations that reduce Fe²⁺-binding in the active site or disrupt LH2 dimer assemblies (Supplementary Fig. iv) without altering LH2 poly peptide levels in 344SQ cells (Supplementary Fig. 5). These mutations reduced orthotopic LUAD metastatic capacity but not size (Fig.2n, o) and did non fully restate the effect of Plod2 KO (Fig.1a-c), suggesting that LH2 drives LUAD progression through tLH-dependent and -contained mechanisms.

LH1 and LH2 have collagen GGT action

With respect to potential tLH-independent mechanisms of action, we reasoned that the GLT domain of LH2 may have enzymatic activity that escaped previous detection²⁷. By sequence alignment, the LH3 GLT domain has 60 and 57% sequence identity to the corresponding domains of LH1 and LH2, respectively, and the DXXD motif²⁸ is strictly conserved in LH1 merely non LH2; D112 and Y114 in LH3 are replaced with glutamate and phenylalanine, respectively, in LH2 to preserve accuse and hydrophobicity (Fig.3a, Supplementary Fig. 6). To make up one's mind whether LH1 and LH2 have galactosylhydroxylysyl glucosyltransferase (GGT) activities, nosotros developed a luciferase assay that detects UDP release following reaction of recombinant LH proteins with UDP-glucose and a constructed amino acid substrate (galactosyl-Hyl) or deglucosylated type 4 collagen. Deglucosylation of blazon Four collagen was achieved past treatment with a collagen glucosidase, protein-glucosylgalactosylhydroxylysine glucosidase (PGGHG) (Supplementary Fig. 7)²⁹. Under these weather condition, all LH family members had detectable GGT activities that were abolished by mutation of Mn²⁺-bounden residues or omission of PGGHG pretreatment (Fig.3b-h).

LH2b is a collagen GGT.

a Amino acid sequence alignment of human LHs. Residues involved in Mn²⁺- and uridine diphosphate (UDP)-binding are indicated with arrows and asterisks, respectively. b and c LH1 galactosylhydroxylysyl glucosyltransferase (GGT) activeness was assayed. Substrates were hyl-gal (b) or blazon IV collagen (c) that had been pre-treated with wild-type (+) or glucosidase-dead mutant (-) protein-glucosylgalactosylhydroxylysine glucosidase (PGGHG). GGT activity was measured past detecting UDP production with an ATP-based luciferase analysis. d–f Wild type and mutant LH2b GGT activity was assayed using an ATP-based luciferase analysis that detects UDP production. Substrates as described in (b) and (c). GGT activity was abolished by mutation of a Mnⁱⁱ⁺-binding residuum (D115E) (f). g and h LH3 GGT assays. Substrates as described in (b) and (c). Residues involved in Mnⁱⁱ⁺-binding were mutated (D112A, D115A) (g). i LH2 protein levels were determined by western blot assay of MC cells stably co-transfected with LH2 shRNA (shLH2) and empty vector (-) or vectors expressing Flag-tagged LH2a or LH2b. Tubulin used as loading control. j Quantification of collagen cantankerous-links in matrices produced by MC cells in (i). HLCC-to-LCC ratio was calculated equally (DHLNL + Pyr)/HHMD. k t-LH assay on recombinant LH proteins (LH2a or LH2b) using trimeric collagen telopeptide as substrate. LH activity was measured by detecting succinate production with an ATP-based luciferase analysis. l and thou LH2a and LH2b GGT activity was measured by an ATP-based luciferase analysis that detects UDP production. Substrates used were hyl-gal (l) or PGGHG-treated type IV collagen (chiliad). north Domain structures of LH2a and LH2b (top). Location of exon 13A-encoded sequences (red bar) in accessory domain (Ac). LH2b homology model was generated using a homology-modelling server SWISS-MODEL (lesser). A recently determined LH3 structure was used equally a template (PDB ID: 6FXT). Exon 13A-encoded loop (arrow). Mn²⁺ (magenta ball) and Feⁱⁱ⁺ (orange balls) in GLT and LH domain active sites, respectively. o LH2'south UDP-glucose-binding affinity was determined past microscale thermophoresis. Fluorescein-conjugated UDP-Glucose (l nM) was titrated with unlike concentrations of LH2a and LH2b recombinant proteins to generate the curves. Curves were used to calculate the M_d values for LH2a (red) and LH2b (cyan). Enzymatic activity assay results are mean values (±S.D.) from triplicate samples (n = 3) and microscale thermophoresis results are hateful values from indistinguishable samples (due north = 2). Error bars indicate ±S.D. p values, 2-tailed Student'due south t examination.

Because key residues in LH3'south GLT active site are but partially conserved in LH2, we speculated that LH2's GGT activeness has a singled-out structural basis. LH2 is alternatively spliced into isoforms that do (LH2b) or do not (LH2a) include exon 13A, which encodes 21 amino acids that are reported to regulate tLH activity^30,31. However, LH2a and LH2b similarly rescued collagen crosslinking defects in LH2-scarce MC cells (Fig.3I, j, Supplementary Figs. viii and 9) and had comparable tLH and hLH activities in enzymatic assays (Fig.3k, Supplementary Fig. 10). In contrast, GGT action was sharply higher in LH2b (Fig.3l, m). To determine how alternative splicing regulates LH2'south GGT activity, we modeled LH2b using the recently determined full-length LH3 construction²⁸ and found that exon 13A adopts a loop conformation that is positioned between α10 and β14 of an accessory domain with a Rossmann fold in shut proximity to the GLT agile site (Fig.3n), which led us to speculate that the loop influences substrate binding analogousness. Past microscale thermophoresis, binding affinity to fluorescein-conjugated UDP-glucose was higher for LH2b than LH2a (Fig.3o), and the isoforms demonstrated distinct binding modes (Supplementary Fig. 11a), suggesting that the exon 13A-encoded loop may enhance collagen GGT activity past operation equally a GLT active site cap. Unlabeled UDP-glucose competed with fluorescein-conjugated UDP-glucose for binding to LH2b with an IC₅₀ of 30 µM (Supplementary Fig. 11b), suggesting that fluorescein is non involved in binding. Thus, LH2b is a collagen GLT that is regulated by cooperative interactions between tandem Rossmann domains.

LH2b drives LUAD growth and metastasis

Glucosylgalactosyl-dihydroxylysinonorleucine (GG-DHLNL) levels were college in man LUAD than they were in adjacent normal lung tissues (Fig.4a), warranting studies to determine how collagen glycosylation is regulated in LUAD and its role in LUAD progression. In The Cancer Genome Atlas lung cancer cohorts, which are mostly early-stage tumors, LH2a is the predominant isoform, whereas normal lung tissues have similarly low levels of the 2 isoforms (Fig.4b-d). In dissimilarity, LH2b levels were equal to or college than LH2a levels in man LUAD prison cell lines and highly, but not poorly, metastatic LUAD prison cell lines derived from K-ras/Tp53-mutant mice (Fig.4e, Supplementary Fig. 12-14)²⁰. Thus, the predominant LH2 isoform switches from LH2a to LH2b during LUAD progression.

LH2b inactivation inhibits orthotopic lung tumor growth and metastasis.

a Quantification of glucosylgalactosyl-DHLNL (GG-DHLNL) in human being LUAD and next normal lung samples (due north = eleven of each). Frozen tissue specimens obtained from Houston Methodist Hospital. b–d LH2a and LH2b mRNA levels in tissue specimens (dots) from The Cancer Genome Atlas accomplice. LUAD (b, n = 516), lung squamous jail cell carcinoma (c, northward = 501), and normal lung (d, northward = 110). e Contrary transcriptase PCR (RT-PCR) analysis of LH2a and LH2b mRNA levels in murine K-ras/Tp53-mutant LUAD cell lines. L32 included every bit loading control. f Quantitative RT-PCR analysis of FOX2 mRNA levels in 344SQ cells transfected with command (siCTL) or FOX2 (#i or #2) siRNAs. Values expressed relative to siCTL-transfected cells (northward = three). g RT-PCR analysis of LH2a and LH2b mRNA levels in cells described in (f). h Numbers of visible primary lung tumors (left plot, due north = ix or ten) and metastases to mediastinal nodes (middle plot, due north = 9 or 10) or contralateral lung (right plot, north = 9 or 10) per mouse (dots). Mice were injected intra-thoracically with parental (WT) or CRISPR/Cas-9-edited 344SQ cells with LH2 exon thirteen A deletion. i Subcutaneous tumor size in mice (dots, n = 10). Tumor area (length times width) determined at time of sacrifice, which was three weeks, 6 weeks and 4 weeks later injection of parental, Δexon 13 A, and L735D 344SQ cells, respectively. j Number of invasion projections per multicellular aggregate (dot) (n = 33 or 40 or 46 or 47). Multicellular aggregates were created in laser-ablated microwells, embedded in collagen, cultured for 3 days, stained, and imaged. Length of each invasive projection (dot) (k, n = 314 or 398 or 562 or 672). Mutants include exon 13 A deletion (Δ13 A) or loss of dimerization (L735D). Projection length in Δexon 13 A aggregates was less than that of parental cells and L735D mutants. l Numbers of visible primary lung tumors (left plot, n = ix or 10) and metastases to mediastinal nodes (centre plot, northward = ix or x) or contralateral lung (right plot, n = 9 or 10) per mouse (dots). Parental (WT) or CRISPR/Cas-nine-edited H358 cells with LH2 exon 13A deletion. Results are mean values (±S.D.) from replicate biological samples. Mistake bars bespeak ±S.D. p values, 2-tailed Student's t test.

Splicing factors that drive exon xiii A inclusion and thereby increase the relative levels of LH2b take been identified^32–34. One of them, FOX2, is of particular interest because information technology regulates alternative splicing driven by epithelial-to-mesenchymal transition (EMT)^35,36, which initiates metastasis in K-ras/Tp53-mutant LUAD models^{twenty,37–39}. Minor interfering RNA-mediated depletion of FOX2 in 344SQ cells decreased LH2b levels (Fig.4f, thousand, Supplementary Fig. fifteen), indicating that FOX2 promotes exon 13 A inclusion in 344SQ cells.

Nosotros subjected 344SQ cells to CRISPR/Cas-ix-mediated deletion of exon 13 A to examine the consequences of LH2b loss on LUAD progression. Relative to parental cells, 344SQ_Δexon thirteen A cells had no detectable change in total LH2 protein levels (Supplementary Fig. v) but demonstrated reduced LH2b and increased LH2a mRNA levels (Supplementary Fig. 16). Orthotopic LUADs and flank tumors generated by 344SQ_Δexon 13A cells in syngeneic, immunocompetent mice were smaller and less metastatic than those generated by parental 344SQ cells (Fig.4h, i). Multicellular aggregates generated by 344SQ_Δexon 13 A cells were less invasive than aggregates generated by parental or tLH-deficient (L735D) 344SQ cells (Fig.4j, one thousand). Orthotopic LUADs in nu/nu mice injected with an exon 13A-deficient H358 man LUAD prison cell line were less metastatic than those generated by parental H358 cells (Fig.4l), but this difference did not reach statistical significance, potentially owing to depression baseline metastatic activity of the orthotopic LUADs or the absenteeism of an intact immune system. The contribution of intratumoral immunity was difficult to define given that the subcutaneous tumors generated past 344SQ_Δexon 13A cells were also minor for flow cytometric analysis. Thus, LH2b's GGT activeness drives LUAD growth and metastasis.

Word

The presence of a fibrotic tumor stroma is correlated with hypoxia, immunosuppression, treatment resistance, and metastasis¹⁴. These features upshot in part from an accumulation of stable collagen cross-links^3,24]. Cancer cells straight the formation of stable collagen cross-links by producing enzymes that hydroxylate (e.1000., LH2) or oxidatively deaminate (e.g., lysyl oxidases) Lys residues on collagen^3,24. Here, nosotros abolished tLH activeness by disrupting either Atomic number 26²⁺-binding in the active site or LH2 dimer associates formation and found that LUAD progression was minimally impaired. Homology modeling and biochemical studies identified collagen GGT activity that results from an alternatively spliced exon in LH2. These findings place a previously unrecognized collagen GGT activity that may enable cancer growth and metastasis.

Collagen glycosylation has a profound impact on collagen fibrillogenesis, cross-link maturation, collagen stability, matrix mineralization, axonal guidance, and platelet activation^xl–44. Glycosylation influences the ability of collagens to function as ligands for receptors (e.g., DDRs, integrins, CD44) that direct cellular functions during embryonic development and tissue repair and are aberrantly expressed in numerous pathologic conditions, including cancer^{iv–half dozen}. Based on our finding that LH1 and LH2 have collagen GGT action, nosotros conclude that all LH family members are bifunctional enzymes that function within an integrated collagen regulatory network, and future studies should readdress polymorphisms and inherited mutations in GLT domains of the PLOD gene family that were previously thought to be non-contributory in individuals with connective tissue disorders^45,46. Moreover, our finding that Hyl glucosylation is a pro-tumorigenic collagen modification opens new avenues of research into how collagen glucosylation influences pro-metastatic processes in the tumor microenvironment.

It remains unclear whether LH2- and LH3-catalyzed GGT activities play distinct or overlapping biochemical and biological roles¹⁸. LH2 and LH3 may glucosylate distinct G-Hyl residues on collagen and/or alter dissimilar collagen types. The finding that LH2'due south GGT action is determined by an alternatively spliced exon raises the possibility that its dual enzymatic functions are coordinately regulated. Such coordinated activities could greatly influence angiogenesis and other LH family-dependent processes in the tumor microenvironment every bit recently reported⁴⁷. Although further studies are necessary to dissect the molecular events that link collagen GGT activity to cancer progression, our data enhance the tantalizing possibility that drugs that target LH2's GLT agile site could be useful for the treatment of LUAD and other LH2-dependent cancer types.

Finally, by performing domain swapping and rescue experiments on LH2-deficient osteoblasts, we showed that LH2'due south LH domain determines its tLH activity. Site-directed mutagenesis and enzymatic activity assays identified residues that are disquisitional for LH2's tLH activity. These results indirectly suggest that a unique arginine cluster near LH2 active site may directly engage acidic residues in collagen telopeptides to facilitate substrate recognition. However, protein crystallographic studies will be required to directly substantiate this possibility.

Methods

Plasmids

Total-length murine PGGHG was cloned into a modified pET-28b (Novagen) vector using NheI and NotI cloning sites with standard PCR-based methods. This modified pET-28b vector has NheI inserted in the linker betwixt His₆ and Thrombin recognition site, which changes the amino acid linker from GS to As. For recombinant poly peptide production in Chinese hamster ovary (CHO) cells, truncated human LH1 (aa22-727) and LH3 (aa32-738) were cloned into BamH1 and Not1 sites of pSGHP1, which was a generous souvenir from Dr. Craig W. Vander Kooi (University of Kentucky). Truncated homo LH2 (residues 33-737 for LH2a and residues 33–758 for LH2b) were cloned into XbaI and Not1 sites of pSGHP1. Betoken mutant constructs were generated using QuickChange Lightning Site-Directed Mutagenesis Kit (Agilent). For ectopic expression in MC cells, murine LH1, LH2a, LH2b and mutant were cloned into XbaI and NotI cloning sites of pEF-bsr with standard PCR-based methods⁴⁸. The identities of all constructs used in this report were confirmed by sequencing. Primers used for cloning and mutagenesis are listed (Supplementary Table one).

CRISPR/Cas-ix Plod2 editing

344SQ and H358 cells were cultured in Roswell Park Memorial Institute 1640 supplemented with 10% FBS (complete media) in a humidified temper with 5% CO₂ at 37 °C. To generate Plod2 KO 344SQ cells, we re-constructed the Cas9-2A-GFP vector (Sigma) to express guide RNAs that flank exon1 of Plod2 nether the U6 promoter (Supplementary Table ii). 344SQ cells were transiently transfected with vector using lipofectamine 2000 transfection reagent (Thermo Fisher). 3 days afterward, the top 5% of GFP⁺ cells were isolated by flow sorting and plated as single cells by limiting dilution method. To screen the clones, genomic DNA was extracted from the cells using QuickExtract^™ DNA Extraction Solution (Epibio, Inc) and amplified by PCR with PCR primers flanking the deleted region. The deletion was confirmed by quantitative RT-PCR analysis of RNA and Western blot analysis of cell lysates.

For D689A and L735D knock-ins, cells were electroporated with 7 µg all-in-ane Cas9/gRNA vector and 0.3 nmol ssODN donor template (Supplementary Tabular array 2). For exon 13 A deletion, cells were electroporated with 10 µg all-in-one Cas9/gRNA vectors. Electroporated cells were allowed to recover for 48 h before beingness sorted for GFP⁺ cells and were then plated by limiting dilution at <1 prison cell per well into 96-well plates in complete medium. Once clones grew to adequate sizes, they were expanded into 24-well plates and so processed to extract mRNA and genomic Dna for PCR amplification. PCR products were and so digested using the SalI enzyme to identify D689A Knock-in clones or digested using the SmaI enzyme to place L735D Knock-in clones or subjected to gel electrophoresis to identify exon 13A-deleted clones. D689A and L735D knock-in clones were further confirmed by Sanger sequencing.

Tumor models

Immunocompetent 129/Sv mice syngeneic to 344SQ cells were bred in-house and randomized to balance the cohorts based on age. Nude mice were purchased from The Academy of Texas MD Anderson Cancer Center ERO section. Mice were placed nether general anesthesia (ketamine/xylazine 50 mg kg⁻¹ and 5 mg kg⁻¹, respectively, delivered by intraperitoneal injection) and an incision was made on the thorax under sterile weather to expose the left lung. Tumor cells (x⁶) were injected directly into the left lung in 50 μl sterile PBS. The incision was closed using staples. The mice were humanely killed afterwards 8 days (344SQ exon 13A-deleted mutants) or 7 days (344SQ LH-inactive mutants) or vii weeks (H358 exon 13A-deleted mutants) post injection. Metastatic tumors visible on the surface of mediastinal lymph nodes or the right lung were manually counted. Investigators were blinded to the cohorts at the time of assessment of metastatic tumor numbers. Mice were excluded from the analysis if they died at the time of tumor cell injection due to hemorrhage or pneumothorax. For period cytometry analysis, syngeneic 129/Sv mice received subcutaneous injections of parental or CRISPR/Cas-9-edited 344SQ cells (ane × 10⁶ per mouse) in the correct flank. The mice were monitored for tumor growth and euthanized 3 weeks afterward the injection or at the first sign of morbidity. They were necropsied to isolate the primary tumors for flow cytometric analysis. Measurements were taken from a single tumor generated in each mouse (n = 8–10 mice per cohort).

Flow cytometry

Tumors were dissociated using the MACS (Miltenyi) mouse tumor dissociation kit. Mechanical dissociation using a gentleMACS Octo dissociator (Miltenyi) was performed followed with enzymatic digestion with collagenase I (0.05% w/v, Sigma), DNase type Iv (xxx U/ml, Sigma), and hyaluronidase type V (0.01% westward/5, Sigma) for 40 min with rocking at 37 °C. Tumor samples were mechanically dissociated again and passed through a 70 μm filter before being stained with fluorochrome-conjugated antibodies in FACS buffer. RBC lysis (Biolegend) was performed on both unmarried cell tumor and splenocytes samples post-obit manufacturer recommendation. Cells were stained for surface markers using fluorochrome-conjugated anti-mouse antibodies (CD45, CD3, CD8, CD4) for 1 h at room temperature. Ghost aqua BV510 (Tonobo) was used to stain dead cells. Cells were fixed using 1% PFA at room temperature for 15 min, then washed twice with perm/wash buffer (Biolegend). Cells were stained with master antibodies at room temperature for 1 h. Cells were filtered and analyzed on BD LSR Fortessa (BD Biosciences) and analyzed using FlowJo software (five.10.v.3; Tree Star). For unmarried color bounty, ultracomp eBeads compensation beads (Thermo Fisher) were used and stained with a single fluorescent-conjugated antibiotic according to manufacturer's instructions. Bounty was calculated automatically using BD FACSDiva 8.0.1. Gating schemes utilized for flow cytometry analysis of T cell subsets and myeloid cells are shown in Supplementary Figs. 17 and eighteen. The antibodies used (clone, dilution, company, catalogue #) are as follows: CD8 PE-Cy7 (53-6.vii) 1/800FCBioLegend/100721, CD3 PE-594 (17A2) 1/100 FCBioLegend/100246, CD4 APC-Cy7 (RM4-five) 1/100 FCBioLegend/100526, FoxP3 PerCp-Cy5.five (FJK-16s) 1/100 FCeBioscience/45-5773-82, CD45 Pacific Bluish (thirty-F11) one/100 FCBioLegend/103126, CD25 BUV395 (PC61) 1/100 FCBD Biosciences/564022, CD11c BV786 (N418) 1/100 FCBioLegend/117335, GR1 BV711 (RB6-8C5) ane/100 FCBioLegend/108443, TIM3 APC (B8.2c12) 1/100 FCBioLegend/134007, PD1 BV605 (29 F.1A12) ane/100 FCBioLegend/135220, F4/fourscore APC (BM8.1) i/100 FCTonbo/20-4801-U100, CD11b BV650 (M1170) one/100 FCBioLegend/101239, Ghost aqua BV510 i/50 FCTonbo.

Cell proliferation

Viable cell densities were quantified in subconfluent culture weather using water-soluble tetrazolium salt ane (WST1) reagent as suggested by manufacturer's instructions (Takara). Measurements were taken from distinct samples. Mean values determined from replicate (north ≥ three) biological samples.

siRNA knockdown

FOX2 siRNAs were purchased from Sigma (SASI_Mm02_00305828 and SASI_Mm02_00305829). 344SQ cells were transfected with 100 nM control siRNAs or siRNAs against FOX2. Total RNAs were extracted from the transfected cells 48 h later and cDNA were generated from the RNA samples using qScript cDNA SuperMix (QuantaBio). The expression levels of PLOD2 isoforms were determined using Real-Fourth dimension PCR.

Multicellular aggregates

Multicellular aggregates were created in a 24-well plate containing 1700 laser-ablated microwells per well as described⁴⁹. The microwells were passivated with 0.05% pluronic acrid for 1 h prior to seeding the cells. 85,000 LUAD cells were seeded per well and cultured for 48 h to generate aggregates, each containing 50 LUAD cells.

To examine invasive projection germination on multicellular aggregates in collagen gels, multicellular aggregates were mixed with rat tail-derived type I collagen to generate collagen gels with defined concentrations (2 mg ml^−ane), volumes (200 µl per gel), and aggregate densities (35 aggregates per gel). The gels were allowed to polymerize upside-down on a glass-bottom 35 mm dish at 37 °C for 30 min. The aggregates were cultured for up to 3 days, fixed and stained with phalloidin for visualization, and imaged with a NikonA1 confocal microscope, 10× objective. Invasive projections were divers as at to the lowest degree one visible LUAD cell protruding out of the amass. The length of the projections was manually quantified using the hand free tool to follow the projection shape (ImageJ). Measurements were taken from distinct samples. Hateful values determined from replicate (due north ≥ 3) biological samples.

Protein expression and purification

PGGHG was expressed in E. coli strain Rosetta (DE3). Cells expressing PGGHG were induced with one mM isopropyl β-D-1-thiogalactopyranoside (IPTG) for 16 h at 16 °C. Cells were collected, pelleted and so resuspended in binding buffer (20 mM Tris, pH 8.0, 200 mM NaCl and fifteen mM imidazole). The cells were lysed by sonication and and then centrifuged at 23,000g for 15 min. The recombinant PGGHG proteins (wild type or D300E inactive mutant) were purified with immobilized metal analogousness chromatography.

Human LH1-three recombinant proteins were purified from CHO cell–derived conditioned medium samples every bit described previously⁵⁰. In brief, LH1-3 recombinant proteins were transiently transfected in new Gibco^™ ExpiCHO^™ cells (Thermo Fisher Scientific, Waltham, MA) with polyethylenimine and expressed as a secreted poly peptide with Due north-terminal His₈ and human growth hormone (hGH) tags via big-calibration suspension culture. The LH1-3–containing conditioned medium samples were harvested by centrifugation at 7000rpm for 10 min, filtered through 0.22 μm EMD Millipore Stericup^™ Sterile Vacuum Filter Units (EMD Millipore, Billerica, MA), concentrated to 100 mL, and buffer-exchanged into Nickel-binding buffer (twenty mM Tris, 200 mM NaCl, 15 mM imidazole, pH eight.0) using the Centramate^™ & Centramate PE Lab Tangential Menstruation Organization (Curtain Life Sciences, Ann Arbor, MI). The recombinant LH proteins were and then purified with tandem immobilized metal affinity chromatography and anion commutation chromatography.

Structure modelling

Human LH2 LH domain and LH2b full length structure homologies were generated by the SWISS-MODEL (Swiss Found of Bioinformatics, Biozentrum, University of Basel, Switzerland) homology server⁵¹. PDB entry 6AX7 and 6FXT were utilized as templates to model LH2 LH domain and LH2b full length, respectively.

LH enzymatic activeness analysis

LH enzymatic activity was measured using a luciferase-based assay as described^l. In brief, the assay was performed in LH reaction buffer (l mM HEPES buffer pH 7.4, 150 mM NaCl) at 37 °C for 1 h with ane μM LH enzymes, 10 μM FeSO4, 100 μM ii-OG, 500 μM ascorbate, ane mM dithiothreitol, 0.01% triton x-100, and one mM wild type (LSYGYDEKSTGGISVP(GPO)₈) or mutant (LSYGYAAKSTGGISVP(GPO)_viii) collagen telopeptide mimics or four μM bovine peel collagen substrate containing no telopeptides (Bovine PureCol^®, Advanced BioMatrix). Collagen telopeptide mimics were dissolved in reaction buffer and incubated overnight at 4 °C to facilitate the formation of trimers, which was confirmed by circular dichroism spectroscopy (Supplementary Fig. 19). Except for LH recombinant poly peptide and bovine skin collagen, all reagents were prepared immediately before apply. All these reagents were dissolved in reaction buffer except for FeSO₄ and collagen, which was prepared in x mM HCl, and the pH of the reaction mixture was checked with pH papers to ensure that HCl did non alter the overall sample pH. Bovine skin collagen was denatured by heating at 95 °C for 5 min and then chilled immediately on ice before use. LH activity was measured past detecting succinate product with an adenosine triphosphate–based luciferase assay (Succinate-Glo^™ JmjC Demethylase/Hydroxylase Assay, Promega, Madison, WI) co-ordinate to manufacturers' instructions. Experiments were performed in triplicate from distinct samples, and an unpaired t-test was used to compare the enzymatic activity of different samples.

Round dichroism spectroscopy

LH2 recombinant proteins or constructed telopeptide mimics were analyzed in 0.01 G sodium phosphate and 150 mM NaCl (pH 7.4) at a concentration of 0.5 mg ml⁻¹. Circular dichroism spectra were measured using a J-810 spectropolarimeter (Jasco, Easton, Dr.) with a two mm path length quartz cuvette. All measurements were performed at 20 °C and three scans averaged for each spectrum. A blank spectrum of phosphate-buffered saline was collected in the same manner and used for background subtraction. Results stand for the mean values from triplicate technical repeats in a unmarried experiment. Each protein was analyzed twice.

Blazon Iv collagen deglucosylation

Man type IV collagen (MilliporeSigma, St. Louis, MO) was deglucosylated in deglucosylation reaction buffer (50 mM acetate buffer pH5.3, 150 mM NaCl) at 37 °C for four h with 100 ug of PGGHG enzymes and 2500 µg type Four collagen. Type Four collagen treated with inactive PGGHG D300E mutant served as a negative command. Later on incubation, the reaction was stopped by incubating at 98 °C for 3 min. The deglucosylated blazon 4 collagen (dgCol4) production was indirectly detected by measuring glucose release using Glucose Colorimetric/Fluorometric Analysis Kit (MilliporeSigma, St. Louis, MO) co-ordinate to manufacturers' instructions. Experiments were performed in indistinguishable from singled-out samples, and an unpaired t-test was used to compare the enzymatic activeness of different samples.

GGT enzymatic activeness assay

GGT activity was measured in reaction buffer (100 mM HEPES buffer pH 8.0, 150 mM NaCl) at 37 °C for i h with 1 μM LH enzymes, 20 μM MnCl₂, 100 μM UDP-glucose (MilliporeSigma, St. Louis, MO), 1 mM dithiothreitol, 0.02% bovine serum albumin, and 1 mM galactosyl hydroxylysine (Cayman Chemical, Ann Arbor, MI) or two μM dgCol4. GGT activity was measured by detecting UDP product with an ATP-based luciferase analysis (UDP-Glo^™ Glycosyltransferase Analysis, Promega, Madison, WI) according to manufacturers' instructions. Experiments were performed in triplicate from distinct samples, and an unpaired t-examination was used to compare the enzymatic activity of different samples.

Western blot

Cells were washed with PBS and lysed with cell lysis buffer (Jail cell Signaling Technology, Danvers, MA) to extract total proteins. Cell lysates were separated past SDS-PAGE, transferred onto nitrocellulose transfer membrane using Trans-Blot Turbo Transfer Arrangement (Bio-Rad), so incubated with master antibodies and horseradish peroxidase-conjugated secondary antibodies (Proteintech, Sigma and GE Healthcare). Poly peptide bands were visualized using Pierce ECL Western blotting substrate (Thermo Fisher Scientific). Experiments were performed in triplicate from distinct samples. Results are representative of replicate experiments.

Microscale thermophoresis

Glucose-UDP-(PEG)₆-Fluorescein Conjugate (10 μl at 50 nM, Sigma) was mixed with equal book of serially diluted unlabeled LH2 proteins in twenty mM HEPES, pH 7.4, 150 mM NaCl, v mM Mn²⁺, 0.05% Tween-twenty. Afterward incubation at 25 °C for 15 min, the samples were loaded into silica capillaries (Nanotemper Technologies). For the competition assay, stock-still concentrations of fluorescein conjugated UDP-Glucose (50 nM) and LH2b (20 µM) were titrated with unlike concentrations of unlabeled UDP-Glucose. Measurements were performed at 20 °C using Monolith NT.115 (Nanotemper Technologies). Information were analyzed (Nanotemper Analysis software. five.1.ii.101) to fit G_d according to the police force of mass action and to determine IC₅₀. The experiment was repeated once. Results stand for the mean values from duplicate biological samples.

Collagen cross-link analyses

For collagen cantankerous-link analysis, MC cells were cultured for two weeks as described. The jail cell/matrix layer was washed with common cold PBS, scraped, collected and pelleted by centrifugation at ten,000rpm for 30 min. The pellets were washed with common cold PBS and distilled water, lyophilized, weighed and aliquoted. Aliquots were reduced with standardized NaB³H₄, acrid hydrolyzed and subjected to amino acid and cross-link analyses as reported⁵². The reducible cantankerous-links, dehydro (deH)-dihydroxylysinonorleucine/its ketoamine, deH-hydroxylysinonorleucine/its ketoamine and deH-histidinohydroxymerodesmosine (for cantankerous-link chemistry, meet²) were analyzed as their reduced forms, i.e., DHLNL, HLNL and HHMD, respectively, and the mature trivalent cantankerous-links pyridinoline and deoxypyridinoline were simultaneously quantified past their fluorescence. All cross-links were quantified as mol/mol of collagen based on the value of 300 residues of hydroxyproline per collagen molecule. The Hyl content in collagen was calculated as Hyl/Hyp Ten 300. Experiments were performed in triplicate from singled-out samples. Mean values determined from replicate (n = three) biological samples. Because the O-glycosidic linkage of the carbohydrate remains intact in base hydrolysis, the glycosylated immature bifunctional cross-link (GG-DHLNL) was analyzed by subjecting human LUAD to base hydrolysis with 2 Northward NaOH and analyzed every bit described previously¹⁸.

Statistics and reproducibility

Measurements were taken from replicate biological samples. Results are representative of replicate experiments. Mean values were determined from replicate (n ≥ iii) biological samples. Statistical significance was determined using 2-tailed Student's t test. Whenever possible, investigators were blinded to the treatment groups at the time of assessment.

Written report approval

All mouse studies were approved past the Institutional Animal Care and Employ Committee at The University of Texas Doc Anderson Cancer Center. The utilize of lung tissues quantification of collagen cross-links in this report was performed under Institutional Review Board–approved protocol IRB(2)0910-01565x at Houston Methodist Inquiry Institute, and written informed consent was obtained from participants or their guardians.

Reporting summary

Farther data on research blueprint is available in the Nature Enquiry Reporting Summary linked to this article.

Supplementary information

Acknowledgements

This work was supported in office by National Institutes of Health grants R01CA105155 (J.M.Thou. and M.Y.) and K99CA225633 (H-F.Thou.); Cancer Prevention and Research Constitute of Texas (CPRIT) grant RP160652 (J.M.Thousand); and the Gloria Lupton Tennison Distinguished Professorship in Lung Cancer (J.1000.K.).

Writer contributions

J.G.K. and H-F.G. conceived the study. J.M.Chiliad. oversaw the work of H-F.Yard., N.B-R., Y.C., Ten.Fifty., J.Y., and X.T. H-F.Thou. designed, performed, and analyzed amino acrid sequence alignment and LH and GLT enzymatic action assays. 1000.T. and One thousand.Y. designed, performed, and analyzed the collagen cantankerous-linking assays. Northward.B-R and P.B. designed, performed, and analyzed the studies on 344SQ multicellular aggregates co-cultured on MC cells. B.L.R. and D.L.Chiliad performed and analyzed the flow cytometry of immune cell populations. Y.C. and X.T. created expression vectors and performed transfections on MC cells. C-L.T., J.A.T, M.D.M., and M.N.P. assisted with interpretation of protein crystallographic analyses that were removed during manuscript revision. M.T-R., R.S., H-F.G., and G.B.F. prepared substrates utilized in enzymatic assays. Ten.Fifty. and J.Y. performed mouse breeding, tumor prison cell injections, and necropsies of orthotopic tumor-bearing mice. 1000.North.D and J.L contributed to the homology model of LH2 and telopeptide that was not included in the paper.

Data availability

Source data underlying plots shown in figures are provided in Supplementary Data ane. All other data, if any, will be available upon reasonable request.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher'due south notation Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

The online version contains supplementary textile available at ten.1038/s42003-021-01982-w.

References

1. Guimarães C. F., Gasperini L., Marques A. P., Reis R. 50. The stiffness of living tissues and its implications for tissue engineering. Nat. Rev. Mater. 5, 351–370 (2020).

2. Yamauchi Thousand, Sricholpech M. Lysine post-translational modifications of collagen. Essays Biochem. 2012;52:113–133. doi: 10.1042/bse0520113. [PMC free commodity] [PubMed] [CrossRef] [Google Scholar]

3. Chen Y, et al. Lysyl hydroxylase two induces a collagen cross-link switch in tumor stroma. J. Clin. Investig. 2015;125:1147–1162. doi: 10.1172/JCI74725. [PMC free commodity] [PubMed] [CrossRef] [Google Scholar]

4. Vogel Due west, Gish GD, Alves F, Pawson T. The discoidin domain receptor tyrosine kinases are activated by collagen. Mol. Cell. 1997;one:thirteen–23. doi: ten.1016/S1097-2765(00)80003-9. [PubMed] [CrossRef] [Google Scholar]

5. Stawikowski MJ, Aukszi B, Stawikowska R, Cudic M, Fields GB. Glycosylation modulates melanoma jail cell alpha2beta1 and alpha3beta1 integrin interactions with type Iv collagen. J. Biol. Chem. 2014;289:21591–21604. doi: 10.1074/jbc.M114.572073. [PMC complimentary article] [PubMed] [CrossRef] [Google Scholar]

6. Lauer-Fields JL, Malkar NB, Richet G, Drauz K, Fields GB. Melanoma cell CD44 interaction with the alpha i(4)1263-1277 region from basement membrane collagen is modulated by ligand glycosylation. J. Biol. Chem. 2003;278:14321–14330. doi: 10.1074/jbc.M212246200. [PubMed] [CrossRef] [Google Scholar]

seven. Scietti L., Campioni M., Forneris F. SiMPLOD, a Construction-Integrated Database of Collagen Lysyl Hydroxylase (LH/PLOD) Enzyme Variants. J. Os Miner. Res. 34, 1376–1382 (2019). [PubMed]

8. Hyland J, et al. A homozygous cease codon in the lysyl hydroxylase gene in two siblings with Ehlers-Danlos syndrome type 6. Nat. Genet. 1992;ii:228–231. doi: x.1038/ng1192-228. [PubMed] [CrossRef] [Google Scholar]

9. Ha-Vinh R, et al. Phenotypic and molecular characterization of Bruck syndrome (osteogenesis imperfecta with contractures of the big joints) caused past a recessive mutation in PLOD2. Am. J. Med. Genet. A. 2004;131:115–120. doi: 10.1002/ajmg.a.30231. [PubMed] [CrossRef] [Google Scholar]

ten. Vahidnezhad H, et al. Mutations in PLOD3, encoding lysyl hydroxylase 3, cause a complex connective tissue disorder including recessive dystrophic epidermolysis bullosa-similar blistering phenotype with abnormal anchoring fibrils and type 7 collagen deficiency. Matrix Biol. 2019;81:91–106. doi: 10.1016/j.matbio.2018.xi.006. [PubMed] [CrossRef] [Google Scholar]

11. Salo AM, et al. A connective tissue disorder caused by mutations of the lysyl hydroxylase three factor. Am. J. Hum. Genet. 2008;83:495–503. doi: 10.1016/j.ajhg.2008.09.004. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

12. van der Slot AJ, et al. Increased formation of pyridinoline cantankerous-links due to college telopeptide lysyl hydroxylase levels is a general fibrotic phenomenon. Matrix Biol. 2004;23:251–257. doi: 10.1016/j.matbio.2004.06.001. [PubMed] [CrossRef] [Google Scholar]

13. Eisinger-Mathason TS, et al. Hypoxia-dependent modification of collagen networks promotes sarcoma metastasis. Cancer Disco. 2013;3:1190–1205. doi: ten.1158/2159-8290.CD-xiii-0118. [PMC gratis article] [PubMed] [CrossRef] [Google Scholar]

14. Yamauchi M, Barker TH, Gibbons DL, Kurie JM. The fibrotic tumor stroma. J. Clin. Investig. 2018;128:16–25. doi: x.1172/JCI93554. [PMC complimentary commodity] [PubMed] [CrossRef] [Google Scholar]

xv. Piersma B., Bank R. A. Collagen cantankerous-linking mediated by lysyl hydroxylase 2: an enzymatic battlefield to combat fibrosis. Essays Biochem. 63, 377–387 (2019). [PubMed]

16. Myllyla R, et al. Expanding the lysyl hydroxylase toolbox: new insights into the localization and activities of lysyl hydroxylase 3 (LH3) J. Prison cell Physiol. 2007;212:323–329. doi: ten.1002/jcp.21036. [PubMed] [CrossRef] [Google Scholar]

17. Schegg B, Hulsmeier AJ, Rutschmann C, Maag C, Hennet T. Core glycosylation of collagen is initiated by two beta(one-O)galactosyltransferases. Mol. Jail cell Biol. 2009;29:943–952. doi: 10.1128/MCB.02085-07. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

18. Sricholpech Thousand, et al. Lysyl hydroxylase 3 glucosylates galactosylhydroxylysine residues in type I collagen in osteoblast culture. J. Biol. Chem. 2011;286:8846–8856. doi: 10.1074/jbc.M110.178509. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

xix. Terajima Chiliad, et al. Role of Glycosyltransferase 25 Domain 1 in Type I Collagen Glycosylation and Molecular Phenotypes. Biochemistry. 2019;58:5040–5051. doi: x.1021/acs.biochem.8b00984. [PMC complimentary article] [PubMed] [CrossRef] [Google Scholar]

20. Gibbons DL, et al. Contextual extracellular cues promote tumor cell EMT and metastasis by regulating miR-200 family unit expression. Genes Dev. 2009;23:2140–2151. doi: 10.1101/gad.1820209. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

21. Jiang H, Hegde Due south, DeNardo DG. Tumor-associated fibrosis as a regulator of tumor amnesty and response to immunotherapy. Cancer Immunol. Immunother. 2017;66:1037–1048. doi: x.1007/s00262-017-2003-i. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

22. Meyaard L. The inhibitory collagen receptor LAIR-i (CD305) J. Leukoc. Biol. 2008;83:799–803. doi: 10.1189/jlb.0907609. [PubMed] [CrossRef] [Google Scholar]

23. Guo HF, et al. Pro-metastatic collagen lysyl hydroxylase dimer assemblies stabilized by Fe(2+)-bounden. Nat. Commun. 2018;9:512. doi: ten.1038/s41467-018-02859-z. [PMC complimentary article] [PubMed] [CrossRef] [Google Scholar]

24. Levental KR, et al. Matrix crosslinking forces tumor progression by enhancing integrin signaling. Prison cell. 2009;139:891–906. doi: 10.1016/j.cell.2009.10.027. [PMC gratis article] [PubMed] [CrossRef] [Google Scholar]

25. Gilkes DM, et al. Procollagen lysyl hydroxylase 2 is essential for hypoxia-induced breast cancer metastasis. Mol. Cancer Res. 2013;11:456–466. doi: 10.1158/1541-7786.MCR-12-0629. [PMC costless article] [PubMed] [CrossRef] [Google Scholar]

26. Gilkes DM, Bajpai S, Chaturvedi P, Wirtz D, Semenza GL. Hypoxia-inducible factor 1 (HIF-1) promotes extracellular matrix remodeling under hypoxic conditions past inducing P4HA1, P4HA2, and PLOD2 expression in fibroblasts. J. Biol. Chem. 2013;288:10819–10829. doi: x.1074/jbc.M112.442939. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

27. Heikkinen J, et al. Lysyl hydroxylase three is a multifunctional protein possessing collagen glucosyltransferase action. J. Biol. Chem. 2000;275:36158–36163. doi: 10.1074/jbc.M006203200. [PubMed] [CrossRef] [Google Scholar]

28. Scietti L, et al. Molecular architecture of the multifunctional collagen lysyl hydroxylase and glycosyltransferase LH3. Nat. Commun. 2018;9:3163. doi: 10.1038/s41467-018-05631-five. [PMC gratuitous article] [PubMed] [CrossRef] [Google Scholar]

29. Hamazaki H, Hamazaki MH. Catalytic site of man poly peptide-glucosylgalactosylhydroxylysine glucosidase: Iii crucial carboxyl residues were adamant past cloning and site-directed mutagenesis. Biochem. Biophys. Res. Commun. 2016;469:357–362. doi: 10.1016/j.bbrc.2015.12.005. [PubMed] [CrossRef] [Google Scholar]

30. Mercer DK, Nicol PF, Kimbembe C, Robins SP. Identification, expression, and tissue distribution of the three rat lysyl hydroxylase isoforms. Biochem. Biophys. Res. Commun. 2003;307:803–809. doi: x.1016/S0006-291X(03)01262-ii. [PubMed] [CrossRef] [Google Scholar]

31. Yeowell HN, Walker LC. Tissue specificity of a new splice form of the human lysyl hydroxylase 2 gene. Matrix Biol. 1999;18:179–187. doi: 10.1016/S0945-053X(99)00013-10. [PubMed] [CrossRef] [Google Scholar]

32. Yeowell HN, Walker LC, Mauger DM, Seth P, Garcia-Blanco MA. TIA nuclear proteins regulate the alternating splicing of lysyl hydroxylase two. J. Investig. Dermatol. 2009;129:1402–1411. doi: 10.1038/jid.2008.386. [PubMed] [CrossRef] [Google Scholar]

33. Venables JP, et al. MBNL1 and RBFOX2 cooperate to establish a splicing programme involved in pluripotent stalk cell differentiation. Nat. Commun. 2013;4:2480. doi: 10.1038/ncomms3480. [PubMed] [CrossRef] [Google Scholar]

34. Braeutigam C, et al. The RNA-binding poly peptide Rbfox2: an essential regulator of EMT-driven alternative splicing and a mediator of cellular invasion. Oncogene. 2014;33:1082–1092. doi: 10.1038/onc.2013.50. [PubMed] [CrossRef] [Google Scholar]

35. Shapiro IM, et al. An EMT-driven alternative splicing programme occurs in homo breast cancer and modulates cellular phenotype. PLoS Genet. 2011;7:e1002218. doi: 10.1371/journal.pgen.1002218. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

36. Schliekelman MJ, et al. Targets of the tumor suppressor miR-200 in regulation of the epithelial-mesenchymal transition in cancer. Cancer Res. 2011;71:7670–7682. doi: 10.1158/0008-5472.Tin can-11-0964. [PMC costless article] [PubMed] [CrossRef] [Google Scholar]

37. Yang Y, et al. The Notch ligand Jagged2 promotes lung adenocarcinoma metastasis through a miR-200-dependent pathway in mice. J. Clin. Investig. 2011;121:1373–1385. doi: x.1172/JCI42579. [PMC free commodity] [PubMed] [CrossRef] [Google Scholar]

38. Chen Fifty, et al. Metastasis is regulated via microRNA-200/ZEB1 axis control of tumour prison cell PD-L1 expression and intratumoral immunosuppression. Nat. Commun. 2014;5:5241. doi: x.1038/ncomms6241. [PMC gratis article] [PubMed] [CrossRef] [Google Scholar]

39. Chen L, et al. CD38-Mediated Immunosuppression as a Mechanism of Tumor Cell Escape from PD-1/PD-L1 Occludent. Cancer Disco. 2018;8:1156–1175. doi: 10.1158/2159-8290.CD-17-1033. [PMC costless article] [PubMed] [CrossRef] [Google Scholar]

40. Banushi B, et al. Regulation of post-Golgi LH3 trafficking is essential for collagen homeostasis. Nat. Commun. 2016;7:12111. doi: 10.1038/ncomms12111. [PMC gratis article] [PubMed] [CrossRef] [Google Scholar]

41. Jurgensen HJ, et al. A novel functional role of collagen glycosylation: interaction with the endocytic collagen receptor uparap/ENDO180. J. Biol. Chem. 2011;286:32736–32748. doi: x.1074/jbc.M111.266692. [PMC free commodity] [PubMed] [CrossRef] [Google Scholar]

42. Sricholpech M, et al. Lysyl hydroxylase 3-mediated glucosylation in type I collagen: molecular loci and biological significance. J. Biol. Chem. 2012;287:22998–23009. doi: 10.1074/jbc.M112.343954. [PMC complimentary article] [PubMed] [CrossRef] [Google Scholar]

43. Schneider VA, Granato M. The myotomal diwanka (lh3) glycosyltransferase and type 18 collagen are critical for motor growth cone migration. Neuron. 2006;50:683–695. doi: x.1016/j.neuron.2006.04.024. [PubMed] [CrossRef] [Google Scholar]

44. Katzman RL, Kang AH, Beachey EH. Collagen-induced platelet aggregation: involement of an agile glycopeptide fragment (alpha1-CB5) Science. 1973;181:670–672. doi: x.1126/science.181.4100.670. [PubMed] [CrossRef] [Google Scholar]

45. Tasker PN, et al. Association of PLOD1 polymorphisms with bone mineral density in a population-based written report of women from the UK. Osteoporos. Int. 2006;17:1078–1085. doi: ten.1007/s00198-006-0129-seven. [PubMed] [CrossRef] [Google Scholar]

46. Mumm S, et al. Bruck syndrome 2 variant defective congenital contractures and involving a novel chemical compound heterozygous PLOD2 mutation. Bone. 2020;130:115047. doi: 10.1016/j.bone.2019.115047. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

47. Goveia J, et al. An Integrated Gene Expression Landscape Profiling Arroyo to Identify Lung Tumor Endothelial Jail cell Heterogeneity and Angiogenic Candidates. Cancer Cell. 2020;37:21–36. doi: 10.1016/j.ccell.2019.12.001. [PubMed] [CrossRef] [Google Scholar]

48. Chen Y, et al. Lysyl Hydroxylase 2 Is Secreted past Tumor Cells and Can Modify Collagen in the Extracellular Infinite. J. Biol. Chem. 2016;291:25799–25808. doi: 10.1074/jbc.M116.759803. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

49. Albritton JL, et al. Ultrahigh-throughput Generation and Label of Cellular Aggregates in Light amplification by stimulated emission of radiation-ablated Microwells of Poly(dimethylsiloxane) RSC Adv. 2016;6:8980–8991. doi: 10.1039/C5RA26022A. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

50. Guo HF, et al. A scalable lysyl hydroxylase two expression organization and luciferase-based enzymatic activity assay. Arch. Biochem. Biophys. 2017;618:45–51. doi: 10.1016/j.abb.2017.02.003. [PMC free commodity] [PubMed] [CrossRef] [Google Scholar]

51. Waterhouse A, et al. SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res. 2018;46:W296–W303. doi: 10.1093/nar/gky427. [PMC complimentary article] [PubMed] [CrossRef] [Google Scholar]

52. Yamauchi M, Terajima M, Shiiba One thousand. Lysine Hydroxylation and Cantankerous-Linking of Collagen. Methods Mol. Biol. 2019;1934:309–324. doi: 10.1007/978-1-4939-9055-9_19. [PubMed] [CrossRef] [Google Scholar]

---

Articles from Communications Biology are provided hither courtesy of Nature Publishing Grouping

---

bergwitteplaidep.blogspot.com

Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8055892/

Share this post