Article
Reference
Human pegivirus persistence in the human blood virome after allogeneic haematopoietic stem cell transplantation
VU CANTERO, Diem-Lan, et al.
Abstract
As commensal viruses are defined by the immunological tolerance afforded to them, any immunomodulation, such as is received during haematopoietic stem cell transplantation, may shift the demarcation between innocuous viral resident and disease-causing pathogen.
VU CANTERO, Diem-Lan, et al . Human pegivirus persistence in the human blood virome after allogeneic haematopoietic stem cell transplantation. Clinical Microbiology and Infection , 2018
PMID : 29787887
DOI : 10.1016/j.cmi.2018.05.004
Available at:
http://archive-ouverte.unige.ch/unige:111768
Disclaimer: layout of this document may differ from the published version.
Original article
Human pegivirus persistence in human blood virome after allogeneic haematopoietic stem-cell transplantation
D.-L. Vu
1,6,*,y, S. Cordey
2,4,y, F. Simonetta
3,y, F. Brito
4,5, M. Docquier
4, L. Turin
2,4, C. van Delden
1,4,6, E. Boely
6, C. Dantin
3, A. Pradier
3, E. Roosnek
4, Y. Chalandon
4,3, E.M. Zdobnov
4,5, S. Masouridi-Levrat
3, L. Kaiser
1,2,41)Division of Infectious Diseases, University of Geneva Hospitals, Geneva, Switzerland
2)Laboratory of Virology, Division of Laboratory Medicine, University of Geneva Hospitals, Geneva, Switzerland
3)Division of Haematology, University of Geneva Hospitals, Geneva, Switzerland
4)Faculty of Medicine, Geneva, Switzerland
5)Swiss Institute of Bioinformatics, Faculty of Medicine, Geneva, Switzerland
6)Swiss Transplant Cohort Study, Basel, Switzerland
a r t i c l e i n f o
Article history:
Received 9 January 2018 Received in revised form 11 April 2018
Accepted 1 May 2018 Available online xxx Editor: G. Antonelli Keywords:
Blood virome Human pegivirus Immune reconstitution Next-generation sequencing Stem-cell transplantation
a b s t r a c t
Objectives: Because commensal viruses are defined by the immunologic tolerance afforded to them, any immunomodulation, such as is received during haematopoietic stem-cell transplantation, may shift the demarcation between innocuous viral resident and disease-causing pathogen.
Methods: We analysed by deep-sequencing the plasma virome of 40 allogeneic haematopoietic stem-cell transplantation patients 1 month after transplantation. Because human pegivirus (HPgV) was highly prevalent, we performed a 1-year screening of 122 plasma samples by specific real-time reverse tran- scription PCR assay. We used the log-rank test and the Gray test to assess association with outcomes, and the Mann-Whitney test and multivariable linear regression model to assess association with T-cell reconstitution.
Results: Polyomaviruses (PyV) (20/40 patients), anelloviruses (16/40), pegiviruses (14/40) and herpes- viruses (14/40) were most frequently identified, including ten cytomegalovirus; three Epstein-Barr virus;
two herpes simplex virus type 1; one human herpesvirus 6b and one human herpesvirus 7; 18 Merkel cellePyV; two BK-PyV; three PyV-6; and one JC-PyV. Papillomavirus and adenovirus were identified in 11 and two patients, respectively. The HPgV specific real-time reverse transcription PCR screening identified 51 of 122 positive samples, high virus loads and persistent infections up to 1 year after transplantation.
Comparison between patients with or without HPgV infection at time of transplantation did not reveal a significant difference in infections, engraftment, survival, graft vs. host disease, relapse or immune reconstitution.
Conclusions: The blood virome after allogeneic haematopoietic stem-cell transplantation includes several DNA viruses, notably herpesviruses and PyV. Among RNA viruses, HPgV is highly prevalent and persists for several months, and it thus may deserve special attention in further research on immune reconstitution.D.-L. Vu, Clin Microbiol Infect 2018;▪:1
©2018 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Introduction
Throughout life, humans encounter countless viruses. While the majority do not cause overt disease, each leaves an indelible mark on the immune system, and some may persist as commensal resi- dents, tolerated under tight immunologic restraint [1]. Defining pathology as the binary presence or absence of a pathogen is
*Corresponding author. D.-L. Vu, Laboratory of Enteric Viruses, Department of Genetics, Microbiology and Statistics, Faculty of Biology, University of Barcelona, Av.
Diagonal 643, 08028 Barcelona, Spain.
E-mail address:diem-lan.vu@ub.edu(D.-L. Vu).
y Thefirst three authors contributed equally to this article, and all should be consideredfirst author.
Contents lists available atScienceDirect
Clinical Microbiology and Infection
j o u r n a l h o m e p a g e :w w w . c l i n i c a l m i c r o b i o l o g y a n d i n f e c t i o n . c o m
https://doi.org/10.1016/j.cmi.2018.05.004
1198-743X/©2018 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
therefore a reductive analysis of a more complex interaction, where immunologic cross-talk from commensal tolerance is able to in- fluence the immune response to pathogens, determine their pa- thology or have an indirect effect on host health [1]. Next- generation sequencing (NGS) can elucidate the true diversity of the human virome, providing a more holistic definition of viral pathogenicity in context of the greater microbiome. The human virome plays a diverse and important role in host biology, either by interacting with its immune system or with other components of its microbiome[2]. Because the virome is essentially composed of a collection of antigens and pathogen-associated molecular patterns, it can influence the host response to other infections[3]or enhance susceptibility to inflammatory diseases[4].
Allogeneic haematopoietic stem-cell transplant (allo-HSCT) re- cipients' virome is of particular interest because of their high state of immunosuppression. The consequences of shifts in immuno- competence towards commensal viruses are still unknown. Inves- tigating the HSCT virome could change the shape of routine viral screening in transplant donors and recipients; there is already ev- idence that complications experienced by HSCT recipients such as
graft vs. host disease may be modulated by the presence of commensal viruses[5].
Using NGS, we analysed the human blood virome of HSCT pa- tients at the peak of immunosuppression and after multiple transfusions. According to the high prevalence of human pegivirus (HPgV) identified and its potential for immune modulation[6], we then more deeply investigated its potential influence on immune reconstitution and patient outcome.
Methods Study design
The primary cohort, composed of 40 plasma specimens of allo- HSCT recipients (Supplementary Table S1), was retrospectively analysed by NGS. Plasma specimens were collected at a median of 33 days (range, 26e40 days) after transplantation (Supplementary Fig. S1).
A specific HPgV investigation included thefirst cohort plus 82 additional allo-HSCT patients. We performed an HPgV specific real-
Table 1
Clinical characteristics of HPgVand HPgVþpatients
Characteristic Value All patients HPgV HPgVþ p
No. of patients 117 78 39
Follow-up (months), median (range)
22 (5e40) 22 (5e39) 26 (10e40) 0.2879
Age (years), median (range) 51 (18e70) 52 (19e70) 50 (18e69) 0.5911
Sex Male 78 (67) 59 (76) 19 (49) 0.0036
Female 39 (33) 19 (24) 20 (51)
Primary disease Acute myeloid leukaemia 49 (42) 26 (33) 23 (59) 0.0001
MDS/MPS/MDPS 33 (28) 25 (32) 8 (21)
Acute lymphoid leukaemia 8 (7) 4 (5) 4 (10)
Lymphoma/chronic lymphoid leukaemia 10 (9) 10 (13) 0 (0)
Chronic myeloid leukaemia 4 (3) 2 (3) 2 (5)
Multiple myeloma 8 (7) 7 (9) 1 (3)
Severe aplastic anaemia 4 (3) 3 (4) 1 (3)
Othera 1 (1) 1 (1) 0 (0)
Disease status at transplant Complete remission 68 (58) 41 (53) 27 (69) 0.0673
Partial remission/progression/relapse 44 (38) 33 (42) 11 (28)
NA 5 (4) 4 (5) 1 (3)
Disease risk index Low 8 (7) 4 (5) 4 (10) 0.198
Intermediate 64 (55) 45 (58) 19 (49)
High 32 (27) 20 (26) 12 (31)
Very high 4 (3) 1 (1) 3 (8)
NA 9 (8) 8 (10) 1 (3)
Conditioning MAC 56 (48) 33 (42) 23 (59) 0.0162
RIC 61 (52) 45 (58) 16 (41)
Donor type Matched sibling 43 (37) 30 (38) 13 (33) 0.1142
Matched unrelated donor 49 (42) 32 (41) 17 (44)
Mismatched relative 14 (12) 11 (14) 3 (8)
Mismatched unrelated donor 11 (9) 5 (6) 6 (15)
Cell source PBSC 100 (85) 66 (85) 34 (87) 0.6836
BM 17 (15) 12 (15) 5 (13)
In vivoTCD No ATG 45 (38) 35 (45) 10 (26) 0.005
ATG 72 (62) 43 (55) 29 (74)
Ex vivoTCD No 78 (67) 56 (72) 22 (56) 0.0184
Yes 39 (33) 22 (28) 17 (44)
CMV status D/R 33 (28) 25 (32) 8 (21) 0.2303
D/Rþ 9 (8) 6 (8) 3 (8)
Dþ/R 17 (15) 12 (15) 5 (13)
Dþ/Rþ 58 (50) 35 (45) 23 (59)
Immunosuppression CNI, MMF 59 (50) 38 (49) 17 (44) 0.300
CNI, MTX 43 (37) 28 (36) 19 (49)
Cy, tacrolimus, MMF 15 (13) 12 (15) 3 (8)
Data are presented asn(%) unless otherwise indicated. Statistical analyses were performed by chi-square or Fisher exact tests for categorical variables and Mann-Whitney test for continuous variables. Myeloid malignancies include acute myeloid leukaemia, MDS/MPS/MDPS and chronic myeloid leukaemia. Lymphoid malignancies include acute lymphoid leukaemia, lymphoma/chronic lymphoid leukaemia and myeloma.
ATG, antithymoglobulin; BM, bone marrow; CMV, cytomegalovirus; CNI, calcineurin inhibitor; Cy, cyclosporine A; D, donor; HPgV, human pegivirus; MAC, myeloablative conditioning; MDS, myelodysplastic syndromes; MMF, mycophenolate mofetil; MPS, myeloproliferative syndrome; MTX, methotrexate; MDPS, myelodysplastic/myelopro- liferative syndrome; NA, not applicable; PBSC, peripheral blood stem cell; R, recipient; RIC, reduced intensity conditioning; TCD, T-cell depletion.
aCategory‘other’includes haemoglobinopathies.
D.-L. Vu et al. / Clinical Microbiology and Infection xxx (2018) 1e8 2
time reverse transcription PCR (rRT-PCR) assay on stored plasma at days 0 and 30 after transplantation for all 122 samples and at four additional time points (day before, thenþ100,þ180 andþ365 days after transplantation), when specimens were available. For the immune reconstitution and outcome analyses, we selected patients receiving transplants for thefirst time (excluding 5/122 patients) and considered the time point day 0 after transplantation to cate- gorize patients as infected (HPgVþ) or noninfected (HPgV).The study was approved by the cantonal ethics committee.
NGS and sequence analysis
Unbiased NGS (DNA and RNA-Seq library preparation paired- end sequencing by using the 100 bp protocol with indexing on a HiSeq 2500 sequencer (Illumina, San Diego, CA, USA) was per- formed on plasma samples. We analysed results using the ezVIR pipeline[7].
HPgV rRT-PCR assay
A total of 190mL of plasma was spiked with 10mL of standardized canine distemper virus of known concentration and extracted with the NucliSENS easyMAG (bioMerieux, Marcy l’Etoile, France) nucleic acid kit in a 25 mL elution volume, according to the
manufacturer's instruction. Extracted RNA was used for HPgV- specific rRT-PCR screening analysis using a previously published assay[8]. PCR assay reaction was performed using the QuantiTect Probe RT-PCR Kit (Qiagen, Germantown, MD, USA) on a StepOne- Plus instrument (Applied Biosystems; Thermo Fisher Scientific, Waltham, MA, USA) under the following cycling conditions: 50C for 30 minutes; 95C for 15 minutes; 45 cycles of 15 seconds at 94C; and 1 minute at 55C. Data were analysed by StepOne 2 (Applied Biosystems; Thermo Fisher Scientific). Analytical sensi- tivity was assessed with a plasmid-derived transcribed RNA including the target region (kindly provided by J. T. Stapleton, Iowa City, IA, USA) and showed a 100% detection limit corresponding to five RNA copies per reaction (approximately 132 RNA copies/mL of plasma).
Flow cytometry
Anticoagulated blood (0.5e1 mL) obtained at 1, 3, 6, 9 and 12 months after transplantation was lysed with ammonium chlo- ride and stained withfluorochrome-conjugated antibodies directed against CD3, CD4, CD8, CD56, CCR7 and CD45RA. Samples were incubated for 15 minutes in the dark, andfluorescence-activated cell sorting (FACS) analysis was performed using a Navios flow cytometer (Beckman Coulter, Brea, CA, USA). FACS data were Fig. 1.NGS analysis. (A) Circular abundance plot of virus families identified by NGS in 40 allo-HSCT patients. Numbers indicate number of patients in whom each of virus families have been identified. (B) Grid plot of viruses identified in 40 allo-HSCT patients. Each line represents one virus and each column one patient's blood sample. Numbers under grid indicate number of distinct viruses identified in each sample. Numbers at right indicate number of patients in whom genome sequences of indicated virus have been identified.
Colours represent approximate number of reads (indicated in legend) matching indicated virus genome detected in each sample. HSCT, haematopoietic stem-cell transplantation;
NGS, next-generation sequencing.
analysed by FlowJo software (Treestar, Ashland, OR, USA). NK cells were identified as CD3CD56þlymphocytes and further divided in CD56brightand CD56dimNK cells. CD3þCD4þand CD3þCD8þT-cell subpopulations were further identified on the basis of CCR7 and CD45RA expression.
Statistical analyses
The chi-square or Fisher's exact tests were used for categorical variables. Mann-Whitney test or Wilcoxon matched pairs signed rank test were used for continuous variables. Log-rank test was used for survival analyses. We used multivariable regression models to adjust for confounding factors (Table 1). Cumulative incidence estimates of acute graft vs. host disease, relapse and in- fections were calculated by the Gray test with death from other causes as a competitive event. p<0.05 was considered statistically significant. Statistics were performed by GraphPad Prism 7 (GraphPad Software, La Jolla, CA, USA), R 3.2.0 with the EZR graphical user interface (R Foundation for Statistical Computing, Vienna, Austria;http://www.r-project.org/) and Stata/IC 13.1 (Sta- taCorp, College Station, TX, USA).
Results
NGS allo-HSCT blood virome study
Among the 40 allo-HSCT patients screened by NGS, three (7.5%) were negative for any virus sequences. Thirty-two (80%) and 19 (47.5%) had at least one detectable DNA and RNA virus sequence, respectively. Among the 37 positive patients, the number of concomitantly detected viruses ranged from one to 5 viruses in a single patient.Fig. 1A provides the virus families identified and the number of patients concerned. Among DNA virus sequences, the most frequent were polyomaviruses (PyV) (20/40 patients, 50%),
followed by anelloviruses (40%), herpesviruses (35%), human pap- illomaviruses (HPV) (27.5%) and adenoviruses (5%). For RNA virus sequences, the most frequent were HPgV (35%), followed by hep- atitis C virus (5%) andRubella virus(2.5%).
Fig. 1B shows the specific virus sequences identified for each patient according to their respective number of reads. The anello- viruses were mostly Torque Teno (TT) virus (15/21); the other cases were TT Midi Virus and TT Mini Virus. Five patients had more than one species detected. Among herpesviruses, cytomegalovirus (CMV) was the most frequent (10/17), followed by Epstein-Barr virus (EBV), herpes simplex (HSV) 1, human herpesvirus (HHV) 6b and HHV-7. Two different types of herpesviruses were simul- taneously detected in three patients. The PyV included Merkel cell (MCPyV) (18/24), BK virus, PyV-6 and JC virus. Four patients were positive for two different PyV. HPV matching to the 107 genotype was predominant among HPV (6/11), followed by genotypes 17, 4, 15 and FA75. No multiple infections were identified for HPV.
There was no differences in the blood virome composition be- tween patients receiving non- and myeloablative conditioning (Mann-Whitney test, data not shown).
Confirmatory testing by PCR
Our r(RT-)PCR-specific assays for HPgV, hepatitis C virus, adenovirus, BK, JC PyV, CMV, EBV and HHV-6 confirmed all cases except two EBV, one BK PyV and one adenovirus. Oppositely, three CMV and one EBV, with relatively low virus loads (<3.210E3 copies/mL) were detected by rPCR but missed by NGS. One BK virus wasfiltered by NGS as a contaminant although then confirmed by rPCR. The HSV-1 and the rubella cases were not confirmed by r(RT-) PCR. We did not screen the HHV-7, the anelloviruses nor the HPV by rPCR.Supplementary Fig. S2provides the HPgV-positive samples phylogenetic tree.
Fig. 2.HPgV rRT-PCR screening. Each box plot represents one time point of screening.Nindicates number of HPgV-positive patients; total, total number of patients screened at indicated time point. Percentage is proportion of positive patients for each time point. BT, before transplantation; HPgV, human pegivirus; rRT-PCR, real-time reverse transcription PCR.
D.-L. Vu et al. / Clinical Microbiology and Infection xxx (2018) 1e8 4
HPgV investigation
Given the high frequency of HPgV detection, the lack of knowledge on its infection kinetics and its potential immuno- modulatory role, we performed an HPgV-specific rRT-PCR screening on 122 consecutive allo-HSCT patients with a mean follow-up of 339 days after transplantation (Fig. 2). Overall, 51 patients (41.8%) were positive for HPgV at least once during the follow-up period. 32% of patients were positive before trans- plantation, of whom all except one patient remained positive until the last time point (days 180e365). In seven of 51 positive patients, the infection appeared after transplantation; we screened the donor serum from 6 of these patients: one donor was positive for HPgV, whose recipient became viraemic at the time of trans- plantation. The median virus load of positive cases ranged between 1.7104and 5.2105copies/mL of plasma (Fig. 2).
HPgV and transplantation outcomes
We next assessed the association between HPgV infection and transplantation outcomes. Table 1 presents the patients' charac- teristics. HPgVþ group included higher proportions of patients with myeloid malignancies (85% vs 68%; p 0.0072) and lower proportions of patients with lymphoid malignancies (13% vs 27%;
p 0.0208). Accordingly, higher proportions of HPgVþ patients received a myeloablative conditioning. No difference was observed in the time to neutrophil engraftment (Fig. 3A), in overall survival, progression-free survival or graft vs. host diseaseefree, relapse- free survival (Fig. 3B), nor in cumulative incidences of grades IIeIV acute graft vs. host disease (Fig. 3C), disease relapse (Fig. 3D) or infections (Supplementary Fig. S3 and Supplementary Table S2 for microbial agents identified) between HPgVþ and HPgV patients.
Fig. 3.Estimates of time to engraftment, survival, aGVHD and relapse in HPgVþand HPgVpatients. (A) Engraftment. (B) Survival: (a) OS; (b) PFS; (c) GRFS. (C) aGvHD. (D) Relapse.
HPgVþpatients had median of 17 days (range, 12e37 days) and HPgVpatients had median of 16 days (range, 12e34 days) before engraftment. OS was calculated from date of allo- HSCT to death or last follow-up. PFS was calculated from date of allo-HSCT until disease relapse or progression or last follow-up. GRFS was calculated from date of allo-HSCT until GRFS-defining event or last follow-up. Events determining GRFS included grade 3/4 aGvHD, chronic GvHD requiring systemic therapy, relapse or death. p values compare HPgVþvs.
HPgVpatients. Percentages indicate outcome at 12 months after transplantation with 95% confidence interval. aGvHD, acute graft vs. host disease; GRFS, graft vs. host dis- easeefree, relapse-free survival; GvHD, graft vs. host disease; HPgV, human pegivirus; OS, overall survival; PFS, progression-free survival; SCT, haematopoietic stem-cell transplantation.
HPgV association with cellular and humoral immunity
We compared cellular immune reconstitution in HPgVþ and HPgV patients over 1 year after transplantation. At 1 and 12 months after transplantation, we detected significantly lower numbers of CD4 and CD8 and higher NK cell counts in HPgVþpa- tients, respectively (Fig. 4A). However, multivariable linear regression analysis, taking into account clinical variables differing between both groups at the studied time points (i.e. sex and disease status at 1 month and sex, donor type and T-cell depletion at 12 months), did not support these differences. Similar analyses on cell subsets distributions among T and NK cells identified lower proportions of CD8 TEMRA T cells in HPgVþpatients at 12 months, although this difference was not confirmed after adjustment (Fig. 4B).
Our analyses also failed to detect any significant impact of intravenous immunoglobulin administration on HPgV titres (Supplementary Fig. S4).
Discussion
We reported all RNA and DNA viruses detectable by unbiased NGS after allo-HSCT. Our results reveal that so-called commensal viruses are a major component of the allo-HSCT blood virome, where HPgV was identified in more than 30% of patients.
Overall, PyV, anellovirus and pegivirus were most frequently identified. The median number of concomitant virus sequences was two per patient, with a maximum offive distinct sequences iden- tified in four patients. According to a previous study, the DNA blood virome of solid organ transplant recipients is composed of 70%
Fig. 4.Immune reconstitution in HPgVþand HPgVpatients from 0 to 12 months after transplantation. (A) Total CD4þ, CD8þand NK cell count. (B) Proportion of distinct subsets within CD4þ, CD8þand NK cells in HPgVor HPgVþpatients. At 1 month, HPgVþpatients had median CD4 count of 0 cells/mL (range, 0e199 cells/mL) and CD8 count of 0 cells/mL (range, 0e478 cells/mL), while HPgVpatients had median CD4 count of 35.31 cells/mL (range, 0e503 cells/mL; p 0.0326) and median CD8 count of 81 cells/mL (range, 0e449 cells/mL;
p 0.0445). At 12 months, HPgVþpatients had higher NK cell counts (median 185 cells/mL, range 50e1090 cells/mL, vs. 125 cells/mL, range 10e460 cells/mL; p 0.0472). HPgVþpatients had lower proportion of CD8 TEMRA T cells (median 26%, range 5e77%; HPgVpatients: median 45%, range 12e93%; p 0.0142). *Comparison of rate of TEMRA cells between HPgVþ and HPgVpatients. CM, central memory; EM, effective memory; HPgV, human pegivirus; NK, natural killer T cells.
D.-L. Vu et al. / Clinical Microbiology and Infection xxx (2018) 1e8 6
Anelloviridae,13% Herpesviridae, 5% Polyomaviridae and 0 Papil- lomaviridae[9]. The equivalentfindings in our cohort is 28%, 22%, 32% and 14%, respectively. This divergence could be partially explained by critical differences in the protocols used for genetic material extraction and bioinformatics pipelines. If comparing with the gut virome of allo-HSCT patients[5], the same persistent vi- ruses than ours were identified (Anelloviridae, Polyomaviridae, HerpesviridaeandPapillomaviridaefound in 70%, 34%, 27% and 18%
of patients, respectively); differences between both studies mainly rely on the viruses' tropism (Flaviviridae is found exclusively in blood and exogenous enteric viruses in the gut).
Anelloviruses are known to cause prolonged viraemia and multiple coinfections. Two thirds of our study population was positive for at least one anellovirus, corresponding to previously reported prevalence[10]. While it has not been linked with overt organ disease, torque teno virus (TTV) has been proposed as a proxy of immunosuppression in both solid organ transplant[11]and allo- HSCT recipients[12].
MCPyV and PyV-6 are found on the skin of healthy individuals.
While MCPyV has been associated with Merkel cell carcinoma, PyV- 6 is currently not associated with any disease. Peripheral blood mononuclear cells (PBMC) are a reservoir for manyPolyomaviridae [13]; we studied cell-free plasma samples, which could explain the relatively low number of reads in the PyVþsamples. Similarly, HPV can persist in PBMCs, and HPV DNA is generally found in 8% to 15%
of healthy blood donors[14], suggesting that HPV can circulate in blood and are not necessarily skin contaminants.
The r(RT-)PCR assays could not confirm the HSV-1 and the rubella results. Clinical history did not indicate any HSV infection and theRubella virusepositive patient was serologically positive before transplantation, as for measles and mumps, suggesting previous vaccination. Although the low number of reads detected in these cases suggests nonspecific sequences, HSV viraemia can occur in up to 26% of critically ill patients[15], including immu- nocompromised patients without organ disease[16]. Moreover, the cumulative burden of double-stranded DNA viruses detected in plasma has been associated with increased mortality in HSCT re- cipients[17], although they rarely caused end-organ disease and were infrequently the direct cause of death. Similarly, in children with primary immunodeficiencies, cases of chronic skin granuloma induced by persistent rubella vaccine strains exist [18]. Thus, commensal viruses may either have an indirect role in pathology or cause unexpected clinical syndrome.
HPgV viraemia is found in 1% to 5% of healthy blood donors in developed countries[6]. Virus load can reach 107copies/mL, and transmission through blood transfusion is well described [19].
HPgV primarily infects haematopoietic stem cells[20]but can also be found in T, B and NK cells and in monocytes[6]. The HPgV genome can persist in serum microvesicles and be further trans- mitted to PBMCs in vitro, leading to active replication and pro- duction of infectious virus particles[8]. The HPgV prevalence we found is similar to that in other studies[21e26](Supplementary Table S3), but our investigation could suggest an HPgV trans- mission by the donor in one patient, although a pretransplantation- undetectable but positive low virus load is not excluded, andfirm evidence are lacking in the absence of sequencing. Previous studies have demonstrated that HPgV can persist for up to 9 years[26]. We provided detailed quantitative viraemia kinetics definitely proving that HPgV is a persistent infection.
Although there is no known HPgV-associated organ disease, HPgV has an immunomodulatory effect in HIV patients[6]. We assessed the potential impact of HPgV on T-cell reconstitution. We identified a difference in the total CD4 and CD8 cell count at 1 month and the rate of CD8 TEMRA at 12 months between HPgV- positive and -negative cases, which could be comparable to what is
seen in HIV patients[27]. The effect did not persist after multi- variable analyses, and our results thus cannot identify any effect of HPgV on cell reconstitution; nevertheless, many confounders can attenuate the potential effect of HPgV, as this was the case in HPgV- coinfected HIV patients who experienced immune reconstitution after highly active antiretroviral therapy[28]. Similarly, our results did not uncover an association between HPgV infection and allo- HSCT outcomes, but we found a significant difference in the pri- mary disease between HPgVþ and HPgV patients, potentially suggesting an association between HPgV and some haematologic malignancies, similar to what we recently reported for TT virus[12].
Our virome NGS study focused on a single time point after transplantation without comparison to the virome before trans- plantation, and we focused our investigation on HPgV because of its immunomodulatory properties, which of interest in the context of allo-HSCT. Yet Legoff et al.[5]have recently demonstrated an in- crease in the proportion of the eukaryotic viruses in the gut virome after allo-HSCT; in addition, picobirnavirus infection was specif- ically predictive of acute graft vs. host disease, reinforcing the need to investigate the role of so-called orphan disease viruses in noninfectious complications.
In conclusion, our study highlights that most of allo-HSCT re- cipients have at least one virus circulating in their blood after transplantation. Our observations that HPgV is predominant and persistent in the allo-HSCT blood virome should promote further research into the potential impact that commensal viruses can have on immunosuppression.
Acknowledgements
We thank J. Schrenzel (Geneva, Switzerland), V. Lazarevic (Geneva, Switzerland) and N. Gaia (Geneva, Switzerland) for their contribution to the bioinformatics analyses, and M.-A. Hartley (Lausanne, Switzerland) for editorial assistance.
Transparency declaration
Supported in part by grants of the Ernst and Lucie Schmidheiny Foundation to LK; the Henri-Dubois-Ferriere Dinu Lipatti Founda- tion to D-LV; and the Faculty of Medicine of Geneva. All authors report no conflicts of interest relevant to this article.
Appendix A. Supplementary data
Supplementary data related to this article can be found at https://doi.org/10.1016/j.cmi.2018.05.004.
References
[1] Vu DL, Kaiser L. The concept of commensal viruses almost 20 years later:
redefining borders in clinical virology. Clin Microbiol Infect 2017;23:688e90.
[2] Virgin HW. The virome in mammalian physiology and disease. Cell 2014;157:
142e50.
[3] Barton ES, White DW, Cathelyn JS, Brett-McClellan KA, Engle M, Diamond MS, et al. Herpesvirus latency confers symbiotic protection from bacterial infec- tion. Nature 2007;447:326e9.
[4] Cadwell K, Patel KK, Maloney NS, Liu TC, Ng AC, Storer CE, et al. Virus-plus- susceptibility gene interaction determines Crohn’s disease gene Atg16L1 phenotypes in intestine. Cell 2010;141:1135e45.
[5] Legoff J, Resche-Rigon M, Bouquet J, Robin M, Naccache SN, Mercier-Delarue S, et al. The eukaryotic gut virome in hematopoietic stem cell transplantation:
new clues in enteric graft-versus-host disease. Nat Med 2017;23:1080e5.
[6] Chivero ET, Stapleton JT. Tropism of human pegivirus (formerly known as GB virus C/hepatitis G virus) and host immunomodulation: insights into a highly successful viral infection. J Gen Virol 2015;96:1521e32.
[7] Petty TJ, Cordey S, Padioleau I, Docquier M, Turin L, Preynat-Seauve O, et al.
Comprehensive human virus screening using high-throughput sequencing with a user-friendly representation of bioinformatics analysis: a pilot study.
J Clin Microbiol 2014;52:3351e61.
[8] Chivero ET, Bhattarai N, Rydze RT, Winters MA, Holodniy M, Stapleton JT.
Human pegivirus RNA is found in multiple blood mononuclear cellsin vivo and serum-derived viral RNA-containing particles are infectiousin vitro. J Gen Virol 2014;95:1307e19.
[9] De Vlaminck I, Khush KK, Strehl C, Kohli B, Luikart H, Neff NF, et al. Temporal response of the human virome to immunosuppression and antiviral therapy.
Cell 2013;155:1178e87.
[10] Maggi F, Bendinelli M. Human anelloviruses and the central nervous system.
Rev Med Virol 2010;20:392e407.
[11] Focosi D, Antonelli G, Pistello M, Maggi F. Torquetenovirus: the human virome from bench to bedside. Clin Microbiol Infect 2016;22:589e93.
[12] Masouridi-Levrat S, Pradier A, Simonetta F, Kaiser L, Chalandon Y, Roosnek E.
Torque teno virus in patients undergoing allogeneic hematopoietic stem cell transplantation for hematological malignancies. Bone Marrow Transplant 2016;51:440e2.
[13] Toracchio S, Foyle A, Sroller V, Reed JA, Wu J, Kozinetz CA, et al. Lympho- tropism of Merkel cell polyomavirus infection, Nova Scotia, Canada. Emerg Infect Dis 2010;16:1702e9.
[14] Chen AC, Keleher A, Kedda MA, Spurdle AB, McMillan NA, Antonsson A. Hu- man papillomavirus DNA detected in peripheral blood samples from healthy Australian male blood donors. J Med Virol 2009;81:1792e6.
[15] Ong DS, Bonten MJ, Spitoni C, Lunel FM, Frencken JF, Horn J, et al. Epidemi- ology of multiple herpes viremia in previously immunocompetent patients with septic shock. Clin Infect Dis 2017;64:1204e10.
[16] Lepiller Q, Sueur C, Solis M, Barth H, Glady L, Lefebvre F, et al. Clinical rele- vance of herpes simplex virus viremia in intensive care unit patients. J Infect 2015;71:93e100.
[17] Hill JA, Mayer BT, Xie H, Leisenring WM, Huang ML, Stevens-Ayers T, et al. The cumulative burden of double-stranded DNA virus detection after allogeneic HCT is associated with increased mortality. Blood 2017;129:2316e25.
[18] Neven B, Perot P, Bruneau J, Pasquet M, Ramirez M, Diana JS, et al. Cutaneous and visceral chronic granulomatous disease triggered by a rubella virus vac- cine strain in children with primary immunodeficiencies. Clin Infect Dis 2017;64:83e6.
[19] Alter HJ, Nakatsuji Y, Melpolder J, Wages J, Wesley R, Shih JW, et al. The incidence of transfusion-associated hepatitis G virus infection and its relation to liver disease. N Engl J Med 1997;336:747e54.
[20] Bailey AL, Lauck M, Mohns M, Peterson EJ, Beheler K, Brunner KG, et al. Du- rable sequence stability and bone marrow tropism in a macaque model of human pegivirus infection. Sci Transl Med 2015;7. 305ra144.
[21] Skidmore SJ, Collingham KE, Harrison P, Neilson JR, Pillay D, Milligan DW.
High prevalence of hepatitis G virus in bone marrow transplant recipients and patients treated for acute leukemia. Blood 1997;89:3853e6.
[22] Corbi C, Traineau R, Esperou H, Ravera N, Portelette E, Benbunan M, et al.
Prevalence and clinical features of hepatitis G virus infection in bone marrow allograft recipients. Bone Marrow Transplant 1997;20:965e8.
[23] Rodriguez-Inigo E, Tomas JF, Gomez-Garcia de Soria V, Bartolome J, Pinilla I, Amaro MJ, et al. Hepatitis C and G virus infection and liver dysfunction after allogeneic bone marrow transplantation: results from a prospective study.
Blood 1997;90:1326e31.
[24] Akiyama H, Nakamura N, Tanikawa S, Sakamaki H, Onozawa Y, Shibayama T, et al. Incidence and influence of GB virus C and hepatitis C virus infection in patients undergoing bone marrow transplantation. Bone Marrow Transplant 1998;21:1131e5.
[25] Ljungman P, Halasz R, Hagglund H, Sonnerborg A, Ringden O. Detection of hepatitis G virus/GB virus C after allogeneic bone marrow transplantation.
Bone Marrow Transplant 1998;22:499e501.
[26] Maruta A, Tanabe J, Hashimoto C, Kato K, Kanamori H, Fukawa H, et al. Long- term liver function of recipients with hepatitis G virus infection after bone marrow transplantation. Bone Marrow Transplant 1999;24:359e63.
[27] Rydze RT, Bhattarai N, Stapleton JT. GB virus C infection is associated with a reduced rate of reactivation of latent HIV and protection against activation- induced T-cell death. Antivir Ther 2012;17:1271e9.
[28] Ernst D, Greer M, Akmatova R, Pischke S, Wedemeyer H, Heiken H, et al.
Impact of GB virus C viraemia on clinical outcome in HIV-1einfected patients:
a 20-year follow-up study. HIV Med 2014;15:245e50.
D.-L. Vu et al. / Clinical Microbiology and Infection xxx (2018) 1e8 8
