- Email: [email protected]
Analysis of human cytomegalovirus strain populations in urine samples of newborns by ultra deep sequencing
Journal of Clinical Virology 73 (2015) 101–104
Contents lists available at ScienceDirect
Journal of Clinical Virology journal homepage: www.elsevier.com/locate/jcv
Short communication
Analysis of human cytomegalovirus strain populations in urine samples of newborns by ultra deep sequencing Irene Görzer a , Slave Trajanoski b , Theresia Popow-Kraupp c , Elisabeth Puchhammer-Stöckl a,∗ a
Department of Virology, Medical University of Vienna, Austria Center for Medical Research, Medical University of Graz, Austria c Division of Clinical Virology, Medical University of Vienna, Austria b
a r t i c l e
i n f o
Article history: Received 17 July 2015 Received in revised form 30 October 2015 Accepted 1 November 2015 Keywords: Cytomegalovirus Congenital infection Genotypes Mixed infection Deep sequencing
a b s t r a c t Background: Different human cytomegalovirus (HCMV) strains may persistently coexist in the human host. In immunosuppressed patients infection with mixed HCMV populations was associated with a more severe course of infection. Congenital HCMV infection may lead to severe fetal disease and possibly mixed HCMV strain infections might have also impact on the clinical consequences for the newborn. Mixed HCMV strain populations were so far detected in saliva but only rarely in urine of congenitally infected newborns. Objectives: We have therefore analyzed the extent of mixed HCMV genotype populations in urine of congenitally infected newborns using a highly sensitive deep sequencing method. Study design: Twenty urine samples (17 initial and 3 follow-up samples) from 17 congenitally infected newborns with a median HCMV DNA load of 7.5 log10 copies/ml were included. Deep sequencing was applied for gO (UL74) genotyping and quantitative real-time PCR assays were used for gB (UL55) and gH (UL75) genotyping. Results: In none of the urine samples a gO genotype mixture was detected, although a mean of 10.000 sequence reads per amplicon was analyzed, which allows to explore gO genotypes down to less than 1% of the total gO sequences. Also only one gB genotype was detected in the patients’ initial samples, while a gH genotype mixture was detected in one case using real time PCR with a sensitivity of 5% for minor populations. Conclusion: Mixed HCMV genotype populations are only rarely found in urine of congenitally infected newborns even when highly sensitive HCMV genotyping methods are applied. © 2015 Elsevier B.V. All rights reserved.
1. Background Human cytomegalovirus (HCMV) infection during pregnancy may lead to congenital infection of the fetus and severe clinical complications in the child [1,2]. HCMV transmission from mother to fetus seems to occur in 30–40% of primary infections, but also HCMV reinfection in presence of pre-existing maternal immunity to HCMV may lead to intrauterine transmission [3,4].Diagnosis of
∗ Corresponding author at: Department of Virology, Medical University Vienna, Kinderspitalgasse 15, A-1090 Vienna, Austria. Fax: +43 1 40160-965599. E-mail addresses: [email protected] (I. Görzer), [email protected] (S. Trajanoski), [email protected] (T. Popow-Kraupp), [email protected] (E. Puchhammer-Stöckl). http://dx.doi.org/10.1016/j.jcv.2015.11.003 1386-6532/© 2015 Elsevier B.V. All rights reserved.
congenital HCMV infection is made by detection of virus in urine, saliva, or blood within the first 2–3 weeks of birth [1]. HCMV establishes a persistent infection over life time, and may reactivate [1]. In addition, reinfections with different HCMV strains may occur [5,6] and multiple HCMV strains may persistently coexist [7]. In transplant patients infections with mixed HCMV strains appear to be associated with a more severe course of infection and disease [8–13]. Thus, also the question to which degree mothers may transmit more than one HCMV strain to the fetus could be clinically relevant. From previous investigations it was suggested that mixed HCMV populations may be frequently detected in saliva of congenitally infected newborns [14,15], while multiple HCMV strains were rarely found in urine of newborns using genotype-specific real-time PCR and Sanger sequencing methods [15–18]. It is, however, unclear whether virus strain mixtures were
102
I. Görzer et al. / Journal of Clinical Virology 73 (2015) 101–104
Table 1 Distribution of HCMV genotypes in urine of 17 congenitally infected newborns. Subject
P3 P12 P16 P4 P5 P6 P11 P15 P1 P7 P9 P2 P10
P13 P18 P8 P14
Age
HCMV DNA load
Genotype
(in days)
(log10 copies/mL)
gOa
gHb
gBb
8 2 16 2 13 12 8 2 98 5 8 3 2 8 32 90 13 18 1 9
6,7 7,5 8,0 7,9 7,6 7,3 5,6 9,9 5,6 7,8 7,2 8,7 9,1 7,0 7,8 8,5 7,8 7,4 4,6 6,3
1a 1a 1a 1b 1b 1b 1b 2a 2a 2a 2a 2b 3 3 3 3 3 3 5 5
1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2c , 1 2 2
1 1 2 1 3 1 1 4 4c , 1 1 1 1 3 1 1 1 1 2 2 3
a Quantitative gO genotyping by ultra-deep pyrosequencing [19]; the sensitivity limit for minor genotypes was 0.1% for patient samples, P6, P9, P10, P11, P14, P15, P18, and 1% for all other samples due to the number of sequence reads per sample. b gB- and gH genotyping by quantitative real-time PCR assays [21]. c Major HCMV genotype (>80% abundance).
undetectable in urine due to the limited sensitivity of previously used methods for minor populations. 2. Objectives
pattern were aligned (ClustalX 1.83), and manually edited (BioEdit 7.0.1). Unique sequence reads were combined into a single gO alignment, from which a neighbor-joining tree was inferred using the Kimura 2-parameter method in MEGA 4.0 [20]. Further, gB- and gH-genotyping was performed by previously described genotype specific real time PCR assays [21]. To investigate potential PCR interferences, HCMV DNA negative urine of neonates spiked with HCMV strains Merlin and AD169 (ratio: 5–5%; total load: 107 and 106 copies/ml) showed accurate quantitation of both genotypes as tested by genotype-specific real-time PCR. The detection limit for minor genotypes in mixtures was 5% for the real-time assays [21] and 0.1–1% for UDPS depending on the number of sequence reads as previously described [19]. 4. Results The overall results obtained by genotype analyses are summarized in Table 1. In all 20 urine samples only one genotype was detected at the gO gene. All gO genotypes except gO1c and gO4 were represented in the subject population, with gO1b and gO3 being the most common genotypes (23.5% each). All individual gO genotype sequences showed 100% identity with reference sequences (Fig. 1). There was no evidence of minor gO genotypes or variants thereof at a frequency of down to 0.1% that coexisted along with the major gO genotype. Next, genotype-specific PCR analyses were performed on the gB and gH gene to further expand the HCMV strain characterization. In the initial urine samples always only one gB genotype was detected. As shown in Table 1, all four major gB genotypes were represented in the patients, with gB1 being the predominant genotype (59%). Also, both gH genotypes were represented and in one case a mixture of both gH genotypes was found at a concentration of 95% for gH2 and 5% for gH1.
We therefore analyzed whether mixed HCMV genotype infections can be found in urine samples of congenitally infected newborns by ultra deep pyrosequencing (UDPS). 3. Study design In the present study all newborns were selected, of whom urine samples were sent to our department between 2001 and 2011 due to suspected congenital HCMV infection and who were diagnosed as congenitally infected by HCMV DNA detection in urine within 18 days of birth. From 17HCMV positive newborns, a total of 20 frozen (–20 ◦ C) stored urine samples were still available, consisting of 17 initial (day 2-day 18 after birth) and 3 follow-up samples (days 32, 90, and 98). The virus loads in original samples significantly correlated with that in the thawed samples when included in the study (Table 1) as identified by Cobas Amplicor CMV Monitor Test (Roche Molecular Systems, Germany). The study was approved by the local ethics committee (EK-number: 1679/2012). HCMV DNA was isolated using the automated NucliSENS® easyMAG® system (BioMérieux, France). UDPS gO genotyping was applied to all samples as previously described [19]. Briefly, the HCMV gO gene region (AD169 position 106555–106853) was amplified using primers composed of 454 adaptors, multiple identifiers and gO-specific sequences. Purified PCR products were quantified, pooled and subjected to UDPS using the GS FLX Titanium sequencing kit in a 454/Roche GS FLX instrument. A mean of 10.611 sequence reads, ranging from 1.998 to 25.154 per sample was achieved after passing the default GS FLX quality filters and after pre-processing using RDP’s pyrosequencing pipeline (http://pyro.cme.msu.edu/). All sequence reads per sample were searched for individual gO genotype signature sequence patterns using self-developed Perl script with regular expressions and all reads that did not match with a well-known gO genotype
Fig. 1. Phylogenetic analysis of all unique gO genotype sequences. Neighbor-joining tree was constructed using the unique gO genotype sequences found in the urine samples of congenitally infected newborns. The tree illustrates the clustering with known prototype genotype sequences (shown in italics).
I. Görzer et al. / Journal of Clinical Virology 73 (2015) 101–104
Follow-up urine samples were tested in two newborns (Table 1). In one of them an additional gB genotype was found at day 98. 5. Discussion Our data show that in nearly all patients (94%) only one HCMV gB-gH-gO genotype is detected in the urine of congenitally infected children within the first 3 weeks of birth. Even by UDPS genotyping, which is highly sensitive for minor genotype populations of down to 0.1% of the total virus population, it appears that in our study cohort the presence of multiple HCMV genotypes in urine after congenital infection is extremely rare. Furthermore, it is unlikely that the rare occurrence of mixed genotypes is due to PCR interferences since HCMV negative urine of neonates spiked with HCMV DNA showed no inhibitory effects. Our data therewith confirm previous findings from two Dutch studies gained with less sensitive Sanger sequencing of PCR-derived amplicons and with genotype-specific real-time PCR assays [17,18]. Only in one of the two studies mixed HCMV genotype populations were detected in 1 of 21 urine samples (4.8%) of congenitally infected newborns. And in the study of Ross et al. [15] only 1 of 13 congenitally HCMV infected children (7.7%) had a mixed HCMV population in urine, as detected by PCR cloning and sequencing and genotype-specific real-time PCR. In our study in one newborn an additional gB genotype appeared in a second urine sample obtained at day 98. It is unclear whether different genotypes were transmitted congenitally and disseminated over time between the compartments or whether the newly emerging genotype resulted from later peri- or postnatal HCMV infections. This raises the question whether urine specimens as only source for HCMV genotyping may underestimate the extent of mixed genotype infections in congenitally infected newborns [15,22] and whether simultaneous testing of different clinical samples might be important to detect the occurrence of mixed genotype infections [14]. However, on the other hand, urine samples may better reflect the genotype profile transmitted during pregnancy while mixed HCMV populations in saliva may also represent later infections during or after birth. Almost all gO, gB and gH genotypes were found in congenitally infected newborns and none of them was particularly predominant, which is in agreement with previous reports [14,15,18,23–27]. The gB genotypes were neither linked with specific gH or gO genotypes, however almost all gH-gO combinations correspond to previously identified linkage groups [25]. Finally, in this study we demonstrate that deep sequencing is useful for HCMV genotype classification in urine samples of congenitally infected newborns, and that in our patient collective mixed HCMV genotypes are only rarely found. Further studies are needed to assess which may be the optimal sample to identify congenitally transmitted HCMV genotypes and which implication the replication of mixed HCMV strains may have on the newborns’ clinical outcome. Competing interests None declared. Funding None. Ethical approval The study was approved by the Institutional Ethics Committee of the Medical University of Vienna (EK-number: 1679/2012).
103
Acknowledgements The authors thank Claudia Kellner and Daniela Böhm for excellent technical support and we are grateful to Gabriele Michelitsch and Ingeborg Klymiuk for performing ultra-deep pyrosequencing.
References [1] E.S. Mocarski, Jr, ST, Pass R.F., Cytomegaloviruses, in: D.M. Knipe DM (ed.), Cytomegalovirus Fields Virology, Lippincott Williams and Wilkins: Philadelphia, 5th ed. 2007, p. 2702–2772. [2] W. Britt, Manifestations of human cytomegalovirus infection: proposed mechanisms of acute and chronic disease, Curr. Top. Microbiol. Immunol. 325 (2008) 417–470. [3] S.B. Boppana, L.B. Rivera, K.B. Fowler, M. Mach, W.J. Britt, Intrauterine transmission of cytomegalovirus to infants of women with preconceptional immunity, N. Engl. J. Med. 344 (2001) 1366–1371. [4] W. Britt, Controversies in the natural history of congenital human cytomegalovirus infection: the paradox of infection and disease in offspring of women with immunity prior to pregnancy, Med. Microbiol. Immunol. 204 (2015) 263–271. [5] Z. Novak, S.A. Ross, R.K. Patro, S.K. Pati, R.A. Kumbla, S. Brice, et al., Cytomegalovirus strain diversity in seropositive women, J. Clin. Microbiol. 46 (2008) 882–886. [6] S.A. Ross, N. Arora, Z. Novak, K.B. Fowler, W.J. Britt, S.B. Boppana, Cytomegalovirus reinfections in healthy seroimmune women, J. Infect. Dis. 201 (2010) 386–389. [7] E. Puchhammer-Stöckl, I. Görzer, Human cytomegalovirus: an enormous variety of strains and their possible clinical significance in the human host, Future Virol. 6 (2011) 259–271. [8] L.F. Lisboa, Y. Tong, D. Kumar, X.L. Pang, A. Asberg, A. Hartmann, et al., Analysis and clinical correlation of genetic variation in cytomegalovirus, Transpl. Infect. Dis. 14 (2012) 132–140. [9] A. Coaquette, A. Bourgeois, C. Dirand, A. Varin, W. Chen, G. Herbein, Mixed cytomegalovirus glycoprotein B genotypes in immunocompromised patients, Clin. Infect. Dis. 39 (2004) 155–161. [10] A. Humar, D. Kumar, C. Gilbert, G. Boivin, Cytomegalovirus (CMV) glycoprotein B genotypes and response to antiviral therapy, in solid-organ-transplant recipients with CMV disease, J. Infect. Dis. 188 (2003) 581–584. [11] K. Ishibashi, T. Tokumoto, K. Tanabe, H. Shirakawa, K. Hashimoto, N. Kushida, et al., Association of the outcome of renal transplantation with antibody response to cytomegalovirus strain-specific glycoprotein H epitopes, Clin. Infect. Dis. 45 (2007) 60–67. [12] O. Manuel, A. Asberg, X. Pang, H. Rollag, V.C. Emery, J.K. Preiksaitis, et al., Impact of genetic polymorphisms in cytomegalovirus glycoprotein B on outcomes in solid-organ transplant recipients with cytomegalovirus disease, Clin. Infect. Dis. 49 (2009) 1160–1166. [13] E. Puchhammer-Stöckl, I. Görzer, A. Zoufaly, P. Jaksch, C.C. Bauer, W. Klepetko, et al., Emergence of multiple cytomegalovirus strains in blood and lung of lung transplant recipients, Transplantation 81 (2006) 187–194. [14] S.K. Pati, S. Pinninti, Z. Novak, N. Chowdhury, R.K. Patro, K. Fowler, et al., Genotypic diversity and mixed infection in newborn disease and hearing loss in congenital cytomegalovirus infection, Pediatr. Infect. Dis. J. 32 (2013) 1050–1054. [15] S.A. Ross, Z. Novak, S. Pati, R.K. Patro, J. Blumenthal, V.R. Danthuluri, et al., Mixed infection and strain diversity in congenital cytomegalovirus infection, J. Infect. Dis. 204 (2011) 1003–1007. [16] R. Arav-Boger, C.B. Foster, J.C. Zong, R.F. Pass, Human cytomegalovirus-encoded alpha -chemokines exhibit high sequence variability in congenitally infected newborns, J. Infect. Dis. 193 (2006) 788–791. [17] J.J. de Vries, E. Wessels, A.M. Korver, A.A. van der Eijk, L.G. Rusman, A.C. Kroes, et al., Rapid genotyping of cytomegalovirus in dried blood spots by multiplex real-time PCR assays targeting the envelope glycoprotein gB and gH genes, J. Clin. Microbiol. 50 (2012) 232–237. [18] J. Nijman, F.S. Mandemaker, M.A. Verboon-Maciolek, S.C. Aitken, A.M. van Loon, L.S. de Vries, et al., Genotype distribution, viral load and clinical characteristics of infants with postnatal or congenital cytomegalovirus infection, PLoS One 9 (2014) e108018. [19] I. Görzer, C. Guelly, S. Trajanoski, E. Puchhammer-Stöckl, Deep sequencing reveals highly complex dynamics of human cytomegalovirus genotypes in transplant patients over time, J. Virol. 84 (2010) 7195–7203. [20] K. Tamura, J. Dudley, M. Nei, S. Kumar, MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0, Mol. Biol. Evol. 24 (2007) 1596–1599. [21] I. Görzer, H. Kerschner, P. Jaksch, C. Bauer, G. Seebacher, W. Klepetko, et al., Virus load dynamics of individual CMV-genotypes in lung transplant recipients with mixed-genotype infections, J. Med. Virol. 80 (2008) 1405–1414. [22] N. Renzette, L. Gibson, B. Bhattacharjee, D. Fisher, M.R. Schleiss, J.D. Jensen, et al., Rapid intrahost evolution of human cytomegalovirus is shaped by demography and positive selection, PLoS Genet. 9 (2013) e1003735.
104
I. Görzer et al. / Journal of Clinical Virology 73 (2015) 101–104
[23] J.F. Bale Jr., J.R. Murph, G.J. Demmler, J. Dawson, J.E. Miller, S.J. Petheram, Intrauterine cytomegalovirus infection and glycoprotein B genotypes, J. Infect. Dis. 182 (2000) 933–936. [24] A.Y. Yamamoto, M.M. Mussi-Pinhata, V.M. de Deus Wagatsuma, L.J. Marin, G. Duarte, L.T. Figueiredo, Human cytomegalovirus glycoprotein B genotypes in Brazilian mothers and their congenitally infected infants, J. Med. Virol. 79 (2007) 1164–1168. [25] H. Yan, S. Koyano, Y. Inami, Y. Yamamoto, T. Suzutani, M. Mizuguchi, et al., Genetic linkage among human cytomegalovirus glycoprotein N (gN) and gO genes, with evidence for recombination from congenitally and post-natally infected Japanese infants, J. Gen. Virol. 89 (2008) 2275–2279.
[26] Z.S. Yu, C.C. Zou, J.Y. Zheng, Z.Y. Zhao, Cytomegalovirus gB genotype and clinical features in Chinese infants with congenital infections, Intervirology 49 (2006) 281–285. [27] E. Paradowska, M. Studzinska, D. Nowakowska, J. Wilczynski, M. Rycel, P. Suski, et al., Distribution of UL144, US28 and UL55 genotypes in Polish newborns with congenital cytomegalovirus infections, Eur. J. Clin. Microbiol. 31 (2012) 1335–1345.
Contents lists available at ScienceDirect
Journal of Clinical Virology journal homepage: www.elsevier.com/locate/jcv
Short communication
Analysis of human cytomegalovirus strain populations in urine samples of newborns by ultra deep sequencing Irene Görzer a , Slave Trajanoski b , Theresia Popow-Kraupp c , Elisabeth Puchhammer-Stöckl a,∗ a
Department of Virology, Medical University of Vienna, Austria Center for Medical Research, Medical University of Graz, Austria c Division of Clinical Virology, Medical University of Vienna, Austria b
a r t i c l e
i n f o
Article history: Received 17 July 2015 Received in revised form 30 October 2015 Accepted 1 November 2015 Keywords: Cytomegalovirus Congenital infection Genotypes Mixed infection Deep sequencing
a b s t r a c t Background: Different human cytomegalovirus (HCMV) strains may persistently coexist in the human host. In immunosuppressed patients infection with mixed HCMV populations was associated with a more severe course of infection. Congenital HCMV infection may lead to severe fetal disease and possibly mixed HCMV strain infections might have also impact on the clinical consequences for the newborn. Mixed HCMV strain populations were so far detected in saliva but only rarely in urine of congenitally infected newborns. Objectives: We have therefore analyzed the extent of mixed HCMV genotype populations in urine of congenitally infected newborns using a highly sensitive deep sequencing method. Study design: Twenty urine samples (17 initial and 3 follow-up samples) from 17 congenitally infected newborns with a median HCMV DNA load of 7.5 log10 copies/ml were included. Deep sequencing was applied for gO (UL74) genotyping and quantitative real-time PCR assays were used for gB (UL55) and gH (UL75) genotyping. Results: In none of the urine samples a gO genotype mixture was detected, although a mean of 10.000 sequence reads per amplicon was analyzed, which allows to explore gO genotypes down to less than 1% of the total gO sequences. Also only one gB genotype was detected in the patients’ initial samples, while a gH genotype mixture was detected in one case using real time PCR with a sensitivity of 5% for minor populations. Conclusion: Mixed HCMV genotype populations are only rarely found in urine of congenitally infected newborns even when highly sensitive HCMV genotyping methods are applied. © 2015 Elsevier B.V. All rights reserved.
1. Background Human cytomegalovirus (HCMV) infection during pregnancy may lead to congenital infection of the fetus and severe clinical complications in the child [1,2]. HCMV transmission from mother to fetus seems to occur in 30–40% of primary infections, but also HCMV reinfection in presence of pre-existing maternal immunity to HCMV may lead to intrauterine transmission [3,4].Diagnosis of
∗ Corresponding author at: Department of Virology, Medical University Vienna, Kinderspitalgasse 15, A-1090 Vienna, Austria. Fax: +43 1 40160-965599. E-mail addresses: [email protected] (I. Görzer), [email protected] (S. Trajanoski), [email protected] (T. Popow-Kraupp), [email protected] (E. Puchhammer-Stöckl). http://dx.doi.org/10.1016/j.jcv.2015.11.003 1386-6532/© 2015 Elsevier B.V. All rights reserved.
congenital HCMV infection is made by detection of virus in urine, saliva, or blood within the first 2–3 weeks of birth [1]. HCMV establishes a persistent infection over life time, and may reactivate [1]. In addition, reinfections with different HCMV strains may occur [5,6] and multiple HCMV strains may persistently coexist [7]. In transplant patients infections with mixed HCMV strains appear to be associated with a more severe course of infection and disease [8–13]. Thus, also the question to which degree mothers may transmit more than one HCMV strain to the fetus could be clinically relevant. From previous investigations it was suggested that mixed HCMV populations may be frequently detected in saliva of congenitally infected newborns [14,15], while multiple HCMV strains were rarely found in urine of newborns using genotype-specific real-time PCR and Sanger sequencing methods [15–18]. It is, however, unclear whether virus strain mixtures were
102
I. Görzer et al. / Journal of Clinical Virology 73 (2015) 101–104
Table 1 Distribution of HCMV genotypes in urine of 17 congenitally infected newborns. Subject
P3 P12 P16 P4 P5 P6 P11 P15 P1 P7 P9 P2 P10
P13 P18 P8 P14
Age
HCMV DNA load
Genotype
(in days)
(log10 copies/mL)
gOa
gHb
gBb
8 2 16 2 13 12 8 2 98 5 8 3 2 8 32 90 13 18 1 9
6,7 7,5 8,0 7,9 7,6 7,3 5,6 9,9 5,6 7,8 7,2 8,7 9,1 7,0 7,8 8,5 7,8 7,4 4,6 6,3
1a 1a 1a 1b 1b 1b 1b 2a 2a 2a 2a 2b 3 3 3 3 3 3 5 5
1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2c , 1 2 2
1 1 2 1 3 1 1 4 4c , 1 1 1 1 3 1 1 1 1 2 2 3
a Quantitative gO genotyping by ultra-deep pyrosequencing [19]; the sensitivity limit for minor genotypes was 0.1% for patient samples, P6, P9, P10, P11, P14, P15, P18, and 1% for all other samples due to the number of sequence reads per sample. b gB- and gH genotyping by quantitative real-time PCR assays [21]. c Major HCMV genotype (>80% abundance).
undetectable in urine due to the limited sensitivity of previously used methods for minor populations. 2. Objectives
pattern were aligned (ClustalX 1.83), and manually edited (BioEdit 7.0.1). Unique sequence reads were combined into a single gO alignment, from which a neighbor-joining tree was inferred using the Kimura 2-parameter method in MEGA 4.0 [20]. Further, gB- and gH-genotyping was performed by previously described genotype specific real time PCR assays [21]. To investigate potential PCR interferences, HCMV DNA negative urine of neonates spiked with HCMV strains Merlin and AD169 (ratio: 5–5%; total load: 107 and 106 copies/ml) showed accurate quantitation of both genotypes as tested by genotype-specific real-time PCR. The detection limit for minor genotypes in mixtures was 5% for the real-time assays [21] and 0.1–1% for UDPS depending on the number of sequence reads as previously described [19]. 4. Results The overall results obtained by genotype analyses are summarized in Table 1. In all 20 urine samples only one genotype was detected at the gO gene. All gO genotypes except gO1c and gO4 were represented in the subject population, with gO1b and gO3 being the most common genotypes (23.5% each). All individual gO genotype sequences showed 100% identity with reference sequences (Fig. 1). There was no evidence of minor gO genotypes or variants thereof at a frequency of down to 0.1% that coexisted along with the major gO genotype. Next, genotype-specific PCR analyses were performed on the gB and gH gene to further expand the HCMV strain characterization. In the initial urine samples always only one gB genotype was detected. As shown in Table 1, all four major gB genotypes were represented in the patients, with gB1 being the predominant genotype (59%). Also, both gH genotypes were represented and in one case a mixture of both gH genotypes was found at a concentration of 95% for gH2 and 5% for gH1.
We therefore analyzed whether mixed HCMV genotype infections can be found in urine samples of congenitally infected newborns by ultra deep pyrosequencing (UDPS). 3. Study design In the present study all newborns were selected, of whom urine samples were sent to our department between 2001 and 2011 due to suspected congenital HCMV infection and who were diagnosed as congenitally infected by HCMV DNA detection in urine within 18 days of birth. From 17HCMV positive newborns, a total of 20 frozen (–20 ◦ C) stored urine samples were still available, consisting of 17 initial (day 2-day 18 after birth) and 3 follow-up samples (days 32, 90, and 98). The virus loads in original samples significantly correlated with that in the thawed samples when included in the study (Table 1) as identified by Cobas Amplicor CMV Monitor Test (Roche Molecular Systems, Germany). The study was approved by the local ethics committee (EK-number: 1679/2012). HCMV DNA was isolated using the automated NucliSENS® easyMAG® system (BioMérieux, France). UDPS gO genotyping was applied to all samples as previously described [19]. Briefly, the HCMV gO gene region (AD169 position 106555–106853) was amplified using primers composed of 454 adaptors, multiple identifiers and gO-specific sequences. Purified PCR products were quantified, pooled and subjected to UDPS using the GS FLX Titanium sequencing kit in a 454/Roche GS FLX instrument. A mean of 10.611 sequence reads, ranging from 1.998 to 25.154 per sample was achieved after passing the default GS FLX quality filters and after pre-processing using RDP’s pyrosequencing pipeline (http://pyro.cme.msu.edu/). All sequence reads per sample were searched for individual gO genotype signature sequence patterns using self-developed Perl script with regular expressions and all reads that did not match with a well-known gO genotype
Fig. 1. Phylogenetic analysis of all unique gO genotype sequences. Neighbor-joining tree was constructed using the unique gO genotype sequences found in the urine samples of congenitally infected newborns. The tree illustrates the clustering with known prototype genotype sequences (shown in italics).
I. Görzer et al. / Journal of Clinical Virology 73 (2015) 101–104
Follow-up urine samples were tested in two newborns (Table 1). In one of them an additional gB genotype was found at day 98. 5. Discussion Our data show that in nearly all patients (94%) only one HCMV gB-gH-gO genotype is detected in the urine of congenitally infected children within the first 3 weeks of birth. Even by UDPS genotyping, which is highly sensitive for minor genotype populations of down to 0.1% of the total virus population, it appears that in our study cohort the presence of multiple HCMV genotypes in urine after congenital infection is extremely rare. Furthermore, it is unlikely that the rare occurrence of mixed genotypes is due to PCR interferences since HCMV negative urine of neonates spiked with HCMV DNA showed no inhibitory effects. Our data therewith confirm previous findings from two Dutch studies gained with less sensitive Sanger sequencing of PCR-derived amplicons and with genotype-specific real-time PCR assays [17,18]. Only in one of the two studies mixed HCMV genotype populations were detected in 1 of 21 urine samples (4.8%) of congenitally infected newborns. And in the study of Ross et al. [15] only 1 of 13 congenitally HCMV infected children (7.7%) had a mixed HCMV population in urine, as detected by PCR cloning and sequencing and genotype-specific real-time PCR. In our study in one newborn an additional gB genotype appeared in a second urine sample obtained at day 98. It is unclear whether different genotypes were transmitted congenitally and disseminated over time between the compartments or whether the newly emerging genotype resulted from later peri- or postnatal HCMV infections. This raises the question whether urine specimens as only source for HCMV genotyping may underestimate the extent of mixed genotype infections in congenitally infected newborns [15,22] and whether simultaneous testing of different clinical samples might be important to detect the occurrence of mixed genotype infections [14]. However, on the other hand, urine samples may better reflect the genotype profile transmitted during pregnancy while mixed HCMV populations in saliva may also represent later infections during or after birth. Almost all gO, gB and gH genotypes were found in congenitally infected newborns and none of them was particularly predominant, which is in agreement with previous reports [14,15,18,23–27]. The gB genotypes were neither linked with specific gH or gO genotypes, however almost all gH-gO combinations correspond to previously identified linkage groups [25]. Finally, in this study we demonstrate that deep sequencing is useful for HCMV genotype classification in urine samples of congenitally infected newborns, and that in our patient collective mixed HCMV genotypes are only rarely found. Further studies are needed to assess which may be the optimal sample to identify congenitally transmitted HCMV genotypes and which implication the replication of mixed HCMV strains may have on the newborns’ clinical outcome. Competing interests None declared. Funding None. Ethical approval The study was approved by the Institutional Ethics Committee of the Medical University of Vienna (EK-number: 1679/2012).
103
Acknowledgements The authors thank Claudia Kellner and Daniela Böhm for excellent technical support and we are grateful to Gabriele Michelitsch and Ingeborg Klymiuk for performing ultra-deep pyrosequencing.
References [1] E.S. Mocarski, Jr, ST, Pass R.F., Cytomegaloviruses, in: D.M. Knipe DM (ed.), Cytomegalovirus Fields Virology, Lippincott Williams and Wilkins: Philadelphia, 5th ed. 2007, p. 2702–2772. [2] W. Britt, Manifestations of human cytomegalovirus infection: proposed mechanisms of acute and chronic disease, Curr. Top. Microbiol. Immunol. 325 (2008) 417–470. [3] S.B. Boppana, L.B. Rivera, K.B. Fowler, M. Mach, W.J. Britt, Intrauterine transmission of cytomegalovirus to infants of women with preconceptional immunity, N. Engl. J. Med. 344 (2001) 1366–1371. [4] W. Britt, Controversies in the natural history of congenital human cytomegalovirus infection: the paradox of infection and disease in offspring of women with immunity prior to pregnancy, Med. Microbiol. Immunol. 204 (2015) 263–271. [5] Z. Novak, S.A. Ross, R.K. Patro, S.K. Pati, R.A. Kumbla, S. Brice, et al., Cytomegalovirus strain diversity in seropositive women, J. Clin. Microbiol. 46 (2008) 882–886. [6] S.A. Ross, N. Arora, Z. Novak, K.B. Fowler, W.J. Britt, S.B. Boppana, Cytomegalovirus reinfections in healthy seroimmune women, J. Infect. Dis. 201 (2010) 386–389. [7] E. Puchhammer-Stöckl, I. Görzer, Human cytomegalovirus: an enormous variety of strains and their possible clinical significance in the human host, Future Virol. 6 (2011) 259–271. [8] L.F. Lisboa, Y. Tong, D. Kumar, X.L. Pang, A. Asberg, A. Hartmann, et al., Analysis and clinical correlation of genetic variation in cytomegalovirus, Transpl. Infect. Dis. 14 (2012) 132–140. [9] A. Coaquette, A. Bourgeois, C. Dirand, A. Varin, W. Chen, G. Herbein, Mixed cytomegalovirus glycoprotein B genotypes in immunocompromised patients, Clin. Infect. Dis. 39 (2004) 155–161. [10] A. Humar, D. Kumar, C. Gilbert, G. Boivin, Cytomegalovirus (CMV) glycoprotein B genotypes and response to antiviral therapy, in solid-organ-transplant recipients with CMV disease, J. Infect. Dis. 188 (2003) 581–584. [11] K. Ishibashi, T. Tokumoto, K. Tanabe, H. Shirakawa, K. Hashimoto, N. Kushida, et al., Association of the outcome of renal transplantation with antibody response to cytomegalovirus strain-specific glycoprotein H epitopes, Clin. Infect. Dis. 45 (2007) 60–67. [12] O. Manuel, A. Asberg, X. Pang, H. Rollag, V.C. Emery, J.K. Preiksaitis, et al., Impact of genetic polymorphisms in cytomegalovirus glycoprotein B on outcomes in solid-organ transplant recipients with cytomegalovirus disease, Clin. Infect. Dis. 49 (2009) 1160–1166. [13] E. Puchhammer-Stöckl, I. Görzer, A. Zoufaly, P. Jaksch, C.C. Bauer, W. Klepetko, et al., Emergence of multiple cytomegalovirus strains in blood and lung of lung transplant recipients, Transplantation 81 (2006) 187–194. [14] S.K. Pati, S. Pinninti, Z. Novak, N. Chowdhury, R.K. Patro, K. Fowler, et al., Genotypic diversity and mixed infection in newborn disease and hearing loss in congenital cytomegalovirus infection, Pediatr. Infect. Dis. J. 32 (2013) 1050–1054. [15] S.A. Ross, Z. Novak, S. Pati, R.K. Patro, J. Blumenthal, V.R. Danthuluri, et al., Mixed infection and strain diversity in congenital cytomegalovirus infection, J. Infect. Dis. 204 (2011) 1003–1007. [16] R. Arav-Boger, C.B. Foster, J.C. Zong, R.F. Pass, Human cytomegalovirus-encoded alpha -chemokines exhibit high sequence variability in congenitally infected newborns, J. Infect. Dis. 193 (2006) 788–791. [17] J.J. de Vries, E. Wessels, A.M. Korver, A.A. van der Eijk, L.G. Rusman, A.C. Kroes, et al., Rapid genotyping of cytomegalovirus in dried blood spots by multiplex real-time PCR assays targeting the envelope glycoprotein gB and gH genes, J. Clin. Microbiol. 50 (2012) 232–237. [18] J. Nijman, F.S. Mandemaker, M.A. Verboon-Maciolek, S.C. Aitken, A.M. van Loon, L.S. de Vries, et al., Genotype distribution, viral load and clinical characteristics of infants with postnatal or congenital cytomegalovirus infection, PLoS One 9 (2014) e108018. [19] I. Görzer, C. Guelly, S. Trajanoski, E. Puchhammer-Stöckl, Deep sequencing reveals highly complex dynamics of human cytomegalovirus genotypes in transplant patients over time, J. Virol. 84 (2010) 7195–7203. [20] K. Tamura, J. Dudley, M. Nei, S. Kumar, MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0, Mol. Biol. Evol. 24 (2007) 1596–1599. [21] I. Görzer, H. Kerschner, P. Jaksch, C. Bauer, G. Seebacher, W. Klepetko, et al., Virus load dynamics of individual CMV-genotypes in lung transplant recipients with mixed-genotype infections, J. Med. Virol. 80 (2008) 1405–1414. [22] N. Renzette, L. Gibson, B. Bhattacharjee, D. Fisher, M.R. Schleiss, J.D. Jensen, et al., Rapid intrahost evolution of human cytomegalovirus is shaped by demography and positive selection, PLoS Genet. 9 (2013) e1003735.
104
I. Görzer et al. / Journal of Clinical Virology 73 (2015) 101–104
[23] J.F. Bale Jr., J.R. Murph, G.J. Demmler, J. Dawson, J.E. Miller, S.J. Petheram, Intrauterine cytomegalovirus infection and glycoprotein B genotypes, J. Infect. Dis. 182 (2000) 933–936. [24] A.Y. Yamamoto, M.M. Mussi-Pinhata, V.M. de Deus Wagatsuma, L.J. Marin, G. Duarte, L.T. Figueiredo, Human cytomegalovirus glycoprotein B genotypes in Brazilian mothers and their congenitally infected infants, J. Med. Virol. 79 (2007) 1164–1168. [25] H. Yan, S. Koyano, Y. Inami, Y. Yamamoto, T. Suzutani, M. Mizuguchi, et al., Genetic linkage among human cytomegalovirus glycoprotein N (gN) and gO genes, with evidence for recombination from congenitally and post-natally infected Japanese infants, J. Gen. Virol. 89 (2008) 2275–2279.
[26] Z.S. Yu, C.C. Zou, J.Y. Zheng, Z.Y. Zhao, Cytomegalovirus gB genotype and clinical features in Chinese infants with congenital infections, Intervirology 49 (2006) 281–285. [27] E. Paradowska, M. Studzinska, D. Nowakowska, J. Wilczynski, M. Rycel, P. Suski, et al., Distribution of UL144, US28 and UL55 genotypes in Polish newborns with congenital cytomegalovirus infections, Eur. J. Clin. Microbiol. 31 (2012) 1335–1345.