KI and WU Polyomaviruses and CD4+ Cell Counts in HIV-1–infected Patients, Italy

To investigate an association between KI and WU polyomavirus (KIPyV and WUPyV) infections and CD4+ cell counts, we tested HIV-1–positive patients and blood donors. No association was found between cell counts and virus infections in HIV-1–positive patients. Frequency of KIPyV infection was similar for both groups. WUPyV was more frequent in HIV-1–positive patients.

B K and JC polyomaviruses are known to infect humans (1,2). Recently, the novel KI polyomavirus (KIPyV) and WU polyomavirus (WUPyV) have been identifi ed in respiratory secretions of children with signs of acute respiratory disease (3,4). However, there is little evidence that these viruses are the causative agents of respiratory disease. The pathogenic role of these viruses in immunocompromised patients is also unclear.
In a study that investigated human polyomaviruses in autopsy lymphoid tissue samples from patients who were positive for HIV, KIPyV was detected in 7.1% of immunocompromised patients with AIDS and in 1.8% of nonimmunocompromised controls; WUPyV was detected in 9.5% of patients with AIDS but not in controls (5). We detected KIPyV and WUPyV in 3.2% and 1.6%, respectively, of plasma samples from HIV-1-infected patients (6). To determine an association between infection with KIPyV and WUPyV and CD4+ cell counts, we obtained plasma samples from HIV-1-positive patients having high and low CD4+ cell counts and a group of healthy controls and tested them for these 2 polyomaviruses.

The Study
Plasma specimens from 153 HIV-1-infected persons (75% male patients, median age 41.9 years, interquartile range 33.8-47.3 years) with high (110 persons) and low (43 persons) CD4+ counts and from 130 blood donors (80% male donors, median age 41 years, interquartile range 32-47.5 years) were obtained at the Foundation Polyclinic Tor Vergata in Rome, Italy, during 2004-2009. Of 153 HIV-1-infected patients, 74 were receiving highly active antiretroviral therapy: a nucleoside reverse transcriptase inhibitor (NRTI) and a protease inhibitor (PI) (n = 35 patients); an integrase inhibitor (INI), an NRTI, and a PI (n = 7); a nonnucleoside-reverse transcriptase inhibitor and an NRTI (n = 26); an INI and an NRTI (n = 2); an INI and an NNRTI (n = 2); a chemokine receptor type 5 antagonist, an NRTI, and a PI (n = 1); and a chemokine receptor type 5 antagonist, an INI, and a nonnucleosidereverse transcriptase inhibitor (n = 1). Sixty patients did not receive any therapy. No information was available for 19 patients. Additional information available for patients included HIV-1 viremia and co-infection with hepatitis B virus and hepatitis C virus.
Phylogenetic analysis of the small T antigen gene of KIPyV and WUPyV was performed as described (6,7). GenBank accession numbers of the sequences used in this analysis are shown in the online Appendix Table (www. cdc.gov/EID/content/16/9/1482-appT.htm).
Total DNA was extracted from 0.2-mL plasma samples by using QIAamp DNA Mini Kit (QIAGEN, Milan, Italy) according to the manufacturer's instructions and stored at -80°C until analysis. Amplifi cation of KIPyV and WUPyV was conducted as described (8,9). A standard curve was created in a 4-log range by using 1:10 serial dilutions of a virus-specifi c standard. The dynamic range was determined by using 10-fold dilutions (10 10 -100 copies/reaction) of each sample. Sensitivity of the 2 methods, which corresponded to the lowest plasma dilution detectable at a frequency of 100%, was evaluated. The dynamic range was 10 2 -10 10 for KIPyV and 10 1 -10 10 for WUPyV.
Statistical analysis was performed by using Epi Info version 3.5.1 software (Centers for Disease Control and Prevention, Atlanta, GA, USA). Odds ratios were determined for associations between infection with HIV and infection with KIPyV and WUPyV and other variables. Statistical signifi cance was assessed by calculating 95% confi dence intervals (CIs) and by using standard nonparametric statistics.
Phylogenetic analysis showed that all WUPyVs identifi ed in this study except WUV-IT4 are closely related to the WUV-IT3 strain identifi ed in an HIV-1 patient (6) (Figure). The KIPyV strains identifi ed are relatively distant from those identifi ed in another study (6), except for strain KIV-RM23, which clusters with KIV-RM21 (6) (Figure).

Conclusions
KIPyV and WUPyV have been identifi ed in respiratory secretions of pediatric patients (3,4). New polyomaviruses have also been detected in immunocompromised patients (10)(11)(12)(13). However, the pathogenic role of these polyomaviruses in immunocompromised patients is unclear. No associations were found between CD4+ cell counts in HIV-1-positive patients and infection with KIPyV or WUPyV. Frequency of KIPyV infection for HIV-1-positive patients was similar to that for blood donors. However, frequency of WUPyV infection was higher for HIV-1-positive patients than for blood donors, although this difference showed borderline signifi cance.
Detection of WUPyV did not show a correlation with virus load for HIV-1 or lower CD4+ cell counts. Seroprevalence of KIPyV and WUPyV in an adult population was 55% and 69%, respectively (14). The higher rate of infection for WUPyV may account for the higher rate of detection for WUPyV in our study population. In a previous study (6), prevalence of WUPyV in plasma of HIV-1 -positive patients was lower than that in our study. This difference may have been caused by the larger sample size in our study.
Phylogenetic analysis did not suggest circulation of specifi c KIPyV and WUPyV strains in HIV-1-positive patients. The KIPyVs identifi ed in this study cluster with those identifi ed in HIV-1-positive patients, and the WUPyVs identifi ed are closely related to the strain identifi ed previously in an HIV-1-positive patient (6). However,  these WUPyV strains also cluster with strain WUV-IT1 and 2 strains identifi ed in stool samples (10).
Our study design and the complex nature of AIDSrelated disease do not enable one to make defi nitive conclusions on the role of novel polyomaviruses in HIV-1-positive patients. However, our data seem to exclude an active role for KIPyV and WUPyV in HIV-1-positive patients.
We detected WUPyV and KIPyV in healthy persons and immunocompromised persons. BK and JC polyomaviruses persist in peripheral blood mononuclear cells in healthy persons (15). However, frequency of detection may vary from 0% to 90% of persons tested. This large variation may refl ect recent infection or virus reactivation in a subgroup of persons (15). Thus, detection of KIPyV and WUPy in blood cells of immunocompetent persons is needed to identify a possible hematologic reservoir. Figure. Maximum likelihood phylogenetic analysis of KI polyomavirus (KIPyV) and WU polyomavirus (WUPyV) small T antigen sequences. Strains identifi ed in this study are in boldface. The tree was rooted by using the midpoint rooting method. Branch lengths were estimated by using the best fi tting nucleotide substitution (Hasegawa, Kishino, and Yano) model according to a hierarchical likelihood ratio test (6,7) and were drawn to scale. Scale bar indicates 0.8 nt substitutions per site. Asterisks along the branches indicate signifi cant statistical support for the clade subtending that branch (p<0.001 by the zero-branchlength test and bootstrap support >65%).