Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Oct 15;93(21):e00843-19.
doi: 10.1128/JVI.00843-19. Print 2019 Nov 1.

Human Cytomegalovirus pUL37x1 Is Important for Remodeling of Host Lipid Metabolism

Yuecheng Xi  1 , Samuel Harwood  1   2 , Lisa M Wise  1 , John G Purdy  3   4
Affiliations

Human Cytomegalovirus pUL37x1 Is Important for Remodeling of Host Lipid Metabolism

Yuecheng Xi et al. J Virol. .

Abstract

Human cytomegalovirus (HCMV) replication requires host metabolism. Infection alters the activity in multiple metabolic pathways, including increasing fatty acid elongation and lipid synthesis. The virus-host interactions regulating the metabolic changes associated with replication are essential for infection. While multiple host factors, including kinases and transcription factors, important for metabolic changes that occur following HCMV infection have been identified, little is known about the viral factors required to alter metabolism. In this study, we tested the hypothesis that pUL37x1 is important for the metabolic remodeling that is necessary for HCMV replication using a combination of metabolomics, lipidomics, and metabolic tracers to measure fatty acid elongation. We observed that fibroblast cells infected with wild-type (WT) HCMV had levels of metabolites similar to those in cells infected with a mutant virus lacking the UL37x1 gene, subUL37x1. However, we found that relative to WT-infected cells, subUL37x1-infected cells had reduced levels of two host proteins that were previously demonstrated to be important for lipid metabolism during HCMV infection: fatty acid elongase 7 (ELOVL7) and the endoplasmic reticulum (ER) stress-related kinase PERK. Moreover, we observed that HCMV infection results in an increase in phospholipids with very-long-chain fatty acid tails (PL-VLCFAs) that contain 26 or more carbons in one of their two tails. The levels of many PL-VLCFAs were lower in subUL37x1-infected cells than in WT-infected cells. Overall, we conclude that although pUL37x1 is not necessary for network-wide metabolic changes associated with HCMV infection, it is important for the remodeling of a subset of metabolic changes that occur during infection.IMPORTANCE Human cytomegalovirus (HCMV) is a common pathogen that asymptomatically infects most people and establishes a lifelong infection. However, HCMV can cause end-organ disease that results in death in the immunosuppressed and is a leading cause of birth defects. HCMV infection depends on host metabolism, including lipid metabolism. However, the viral mechanisms for remodeling of metabolism are poorly understood. In this study, we demonstrate that the viral UL37x1 protein (pUL37x1) is important for infection-associated increases in lipid metabolism, including fatty acid elongation to produce very-long-chain fatty acids (VLCFAs). Furthermore, we found that HCMV infection results in a significant increase in phospholipids, particularly those with VLCFA tails (PL-VLCFAs). We found that pUL37x1 was important for the high levels of fatty acid elongation and PL-VLCFA accumulation that occur in HCMV-infected cells. Our findings identify a viral protein that is important for changes in lipid metabolism that occur following HCMV infection.

Keywords: cytomegalovirus; human herpesviruses; lipidomics; metabolism; metabolomics.

PubMed Disclaimer

Figures

FIG 1
FIG 1
subUL37x1 mutant virus infection under serum-free conditions. (A) Western blot analysis reveals that subUL37x1 fails to express pUL37x1, while the expression of neighboring genes is unaffected. (B) Infection in fully confluent fibroblast cells under serum-free conditions was visually tracked from 0.25 to 4 dpi. Images from uninfected control cells at the 0.25- and 4-dpi time points are included. Arrows point to mutant-virus-infected cells that have a morphology (cytopathic effect) similar to that of WT-infected cells. (C) Infectious virus particles released by cells infected with the WT or the subUL37x1 mutant at an MOI of 3 were measured at 4 dpi. (D) The particle-to-infectious unit ratios, as measured by viral DNA and infectious titer levels, were compared for particles released by WT- and subUL37x1-infected cells at 4 dpi, at an MOI of 3. For panels C and D, the error bars are the standard deviations (SD) from three independent experiments (n = 3). *, P = <0.05 to 0.01; **, P < 0.01 (by a t test).
FIG 2
FIG 2
Metabolomic comparison of WT- and subUL37x1-infected cells. (A) Comparison of intracellular metabolite levels in fibroblast cells infected with the WT or mutant subUL37x1 virus relative to those in uninfected cells, at an MOI of 3. (B) Correlation plots for metabolite levels normalized to values for uninfected cells (y axis, WT; x axis, subUL37x1). Correlation R values are given for each time point from 0.25 to 3 dpi. For both panels A and B, metabolites were measured from technical replicates from 4 independent experiments (n = 4).
FIG 3
FIG 3
Concentrations of saturated very-long-chain fatty acids (VLCFAs) in WT- and subUL37x1-infected fibroblasts. FA levels relative to those in uninfected cells are shown for cells infected at an MOI of 3. Shown are FA changes at 2 dpi (A) and at 3 dpi (B). All data are represented as means and SD (n = 4). P values for comparison of mock and subUL37x1 infection that are near the 0.05 cutoff are shown. *, P = <0.05 to 0.01; **, P < 0.01 (by one-way analysis of variance [ANOVA] with a Tukey test).
FIG 4
FIG 4
Synthesis of VLCFAs in WT- and subUL37x1-infected cells. (A) Diagram depicting the flow of carbons from glucose and acetate to FA synthesis and elongation. Carbon atoms are represented by circles; red and orange circles represent 13C-labeled carbons, while black circles depict unlabeled atoms. (B) Incorporation of labeled atoms from [1,6-13C]glucose into saturated FAs. The data are presented as percentages of labeled to unlabeled saturated FAs. (C) Percent labeling of saturated FAs from [1,2-13C]acetate. (D and E) Labeling pattern of C26:0 from labeled glucose (D) and labeled acetate (E). All data are represented as means and SD (n = 3). *, P = <0.05 to 0.01; **, P < 0.01 (by one-way ANOVA with a Tukey test).
FIG 5
FIG 5
ELOVL7 fatty acid elongase protein accumulation during HCMV infection. (A) ELOVL7 protein expression was determined by Western blotting from 2 to 4 dpi. (B) Further independent analysis of ELOVL7 expression at 3 dpi. (C) Quantification of ELOVL7 levels at 3 dpi. ELOVL7 levels were normalized to tubulin levels and are shown relative to WT levels. All data are represented as means and SD (n = 3). *, P < 0.05 (by a t test).
FIG 6
FIG 6
Impact of HCMV infection on proteins involved in the synthesis of VLCFAs. (A) Schematic of the glucose-to-FA synthesis pathway. Enzymes involved in metabolizing glucose carbons to FAs are shown in blue boxes. (B) Western blotting of proteins involved in glucose metabolism and FA synthesis and their regulation. (C) Western blotting at 3 dpi showing ACSS2 expression levels. (D) Quantification of ACSS2 protein levels from three independent experiments. Protein loading was normalized to the tubulin signal. Data are represented as means and SD.
FIG 7
FIG 7
PERK protein levels in WT-infected and subUL37x1-infected cells. (A) Western blotting at 3 dpi showing PERK expression levels in human fibroblast cells infected with WT or subUL37x1 virus at an MOI of 3. (B) Quantification of PERK protein levels from three independent experiments. Protein loading was normalized to the tubulin signal. Data are represented as means and SD. *, P < 0.05 (using a one-sample t test).
FIG 8
FIG 8
Phospholipid levels 3 days after infection with HCMV. (A, left) LC-MS/MS phospholipidomic analysis of WT-infected cells compared to uninfected cells. (Right) The same type of lipidomic analysis showing the levels of phospholipids in subUL37x1-infected relative to WT-infected cells. Both heat maps are organized going from smallest to largest changes for subUL37x1 relative to the WT. (B) Phosphatidylcholine (PC) lipids with VLCFA tails. (C) Similar plots showing phosphatidylserine (PS) lipids. (D) The FA tails in PC and PS lipids shown in panels B and C were identified by MS/MS. In some cases, only one of the two tails was observed. The unobserved tails are marked by ^. All data are from day 3 postinfection for cells infected at an MOI of 3 (n = 3). *, P = <0.05 to 0.01; **, P < 0.01 (by one-way ANOVA with a Tukey test). Box plots show the means plus 25 to 75% variance, while the whiskers show the SD. Individual data points are also shown. PA, phosphatidic acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphatidylinositol; PS, phosphatidyserine; LPC, lysoPC (lysophosphatidylcholine); LPE, lysoPE (lysophosphatidylethanolamine).
FIG 9
FIG 9
Phosphatidylcholine and phosphatidylserine levels in HCMV-infected cells treated with an apoptosis inhibitor. (A) At 3 dpi, cell survival was determined for cells treated with DMSO or 35 μM Nα-tosyl-l-lysine chloromethyl ketone (TLCK). (B) The abundances of PC lipids at 3 dpi were measured in WT- or subUL37x1-infected cells treated with TLCK relative to DMSO. (C) Abundances of PS lipids under the same conditions (n = 3). *, P = <0.05 to 0.01; **, P < 0.01 (by a t test). Box plots show the means plus the 25 to 75% quartile groups, while the whiskers show the SD. Individual data points are also shown.
FIG 10
FIG 10
Levels of lipids in uninfected fibroblast cells expressing pUL37x1. (A) Western blot analysis of uninfected fibroblasts stably expressing pUL37x1. Two independent clones were analyzed and compared to cells stably expressing GFP. (B) PC lipid analysis of pUL37x1-expressing cells relative to GFP control cells. For clarity, the means and standard errors (SE) measured are shown. (C) Box plot of selected PC lipids from panel B. (D) PS lipid analysis of pUL37x1-expressing cells compared to GFP control cells. Shown are the means and SE. (E) Box plot of selected PS lipids (n = 3). *, P < 0.05 (using one-way ANOVA with a Tukey test). Box plots show the means plus the 25 to 75% quartile groups, while the whiskers show the SD. Individual data points are also shown.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Mocarski ES, Shenk T, Griffiths P, Pass RF. 2013. Cytomegaloviruses, p 1960–2014. In Knipe DM, Howley PM, Cohen JI, Griffin DE, Lamb RA, Martin MA, Racaniello VR, Roizman B (ed), Fields virology, 6th ed Lippincott Williams & Wilkins, Philadelphia, PA.
    1. Britt W. 2008. Manifestations of human cytomegalovirus infection: proposed mechanisms of acute and chronic disease. Curr Top Microbiol Immunol 325:417–470. - PubMed
    1. Hyde TB, Schmid DS, Cannon MJ. 2010. Cytomegalovirus seroconversion rates and risk factors: implications for congenital CMV. Rev Med Virol 20:311–326. doi:10.1002/rmv.659. - DOI - PubMed
    1. Britt WJ. 2017. Congenital human cytomegalovirus infection and the enigma of maternal immunity. J Virol 91:e02392-16. doi:10.1128/JVI.02392-16. - DOI - PMC - PubMed
    1. Goldmacher VS, Bartle LM, Skaletskaya A, Dionne CA, Kedersha NL, Vater CA, Han JW, Lutz RJ, Watanabe S, Cahir McFarland ED, Kieff ED, Mocarski ES, Chittenden T. 1999. A cytomegalovirus-encoded mitochondria-localized inhibitor of apoptosis structurally unrelated to Bcl-2. Proc Natl Acad Sci U S A 96:12536–12541. doi:10.1073/pnas.96.22.12536. - DOI - PMC - PubMed

Publication types

MeSH terms