Share this article on:

Human parvovirus 4 in the bone marrow of Italian patients with AIDS

Longhi, Erikaa; Bestetti, Giovannaa; Acquaviva, Veronicaa; Foschi, Antonellaa; Piolini, Robertaa; Meroni, Lucaa; Magni, Carloa; Antinori, Spinelloa; Parravicini, Carlob; Corbellino, Marioa

doi: 10.1097/QAD.0b013e3281e38558
Research Letters

Human parvovirus 4 (PARV4) is a recently discovered member of the Parvoviridae. We investigated the presence of this virus in bone-marrow aspirates of 35 Italian patients with AIDS. Viral DNA was detected by polymerase chain reaction in over 40% of patients (16/35). The infection was most prevalent in injection drug users (IDU; 12/18; 66.7%) as opposed to non-IDU (4/17; 23.5%). PARV4 infection is widespread in Italian patients with AIDS.

aDepartment of Infectious Diseases, Luigi Sacco Hospital, Milan, Italy

bDepartment of Pathology, Luigi Sacco Hospital, Milan, Italy.

Received 3 April, 2007

Accepted 11 April, 2007

Parvovirus 4 (PARV4) is a human virus discovered in 2005 using a polymerase chain reaction (PCR)-based, sequence independent method in the plasma of an HIV-negative homeless injection drug user (IDU) with acute viral infection syndrome [1]. The virus belongs to the Parvoviridae subfamily, but is only distantly related to parvovirus B19 with which it shares less than 30% homology at the amino acid level. Little is known about the epidemiology, clinical correlates of infection, and pathogenic potential of this novel infectious agent.

PARV4 has been detected in 5% (seven of 137) plasma pool samples from north American, but not European, healthy donors [2], in the plasma obtained post mortem of 30% (three of 10) hepatitis C virus (HCV)-infected IDU from the London area [3], and in the bone marrow and lymphoid organs (lymph nodes or spleen) screened after autopsy of 54% (17 of 24) of HIV-positive individuals from Scotland [4]. In the latter study all HIV patients with molecular evidence of PARV4 infection had a history of injection drug use.

In HIV-infected patients, PARV4 viral loads never exceeded 104 copies/106 cells in bone marrow or secondary lymphoid organs, and no association was found between the degree of immunosuppression and viral replication [4]. Similarly, the three HCV-infected IDU described by Fryer et al. [2] had low levels of PARV4 viraemia (e.g. < 700 copies/ml).

Two PARV 4 genotypes are known, PARV4 genotype 1 (e.g. prototype PARV4) and PARV genotype 2 (also called PARV5). These subtypes share approximately 87–92% nucleotide identity over the open reading frame 1 (ORF1) of the virus that is homologous to the non-structural proteins of other parvoviruses [2,4]. Preliminary data show that HIV-seropositive IDU from Scotland who are infected with PARV4 are younger than those infected with PARV5 (median year of birth 1964 and 1953, respectively) [4]. In addition, direct sequence analysis of PCR products reveals 100% identity of PARV4 sequences as opposed to 96.9% for PARV5 [4]. These findings suggest the recent introduction of PARV4 in IDU in the United Kingdom [4], a phenomenon originally described in the general population of Finland for human parvovirus B19 variants [5].

We looked for the presence of PARV4 DNA sequences in a randomly selected series of bone-marrow aspirates obtained from 35 Italian inpatients with AIDS, followed from 1997 to 2005 at the Department of Infectious Diseases, Luigi Sacco Hospital in Milan. Bone marrow aspirates (BMA) were obtained for diagnostic purposes, whenever clinically indicated by the physicians caring for the patients. In addition to traditional culture for standard bacteria, mycobacteria and fungi, a fraction of the BMA was devoted to PCR-based diagnostic techniques (e.g. Leishmania spp., Mycobacterium spp., immunoglobulin gene and T-cell receptor rearrangement DNA PCR) [6]. Excess DNA was stored at −70°C for future studies.

Twenty-six patients were men (74%) and nine were women (26%). A history of drug injection was present in 18 (51%) of them, whereas the remainder were homosexual men (seven of 35; 20%) or reported unprotected heterosexual activities (10 of 35; 29%). The mean age of patients and CD4 T-lymphocyte counts were 44.7 years (range 33–64) and 152 cells/μl (range 2–1200), respectively. Hepatitis C serology was positive in 20 out of 35 patients (57%).

BMA were collected using ethylenediamine tetraacetic acid as an anticoagulant. DNA was extracted using the EZ DNA kit (Invitrogen, Carlsbad, California, USA), and its concentration was measured by spectrophotometry. One microgram of DNA was loaded onto each PCR reaction vessel. PARV4 PCR was performed exactly as described by Fryer et al. [2] using primers PV4OF1F and PV4ORF1R that amplify a 220 basepair region of ORF1. The presence and integrity of human genomic DNA, as well as the absence of major Taq DNA polymerase inhibitors, were assessed using primers that amplify a 252 bp fragment of the human beta-globin gene [7]. PCR products were analysed by 2% agarose gel electrophoresis. In order to estimate the PARV4 load in positive specimens, six serial 10-fold dilutions of the extracted DNA were performed [7]. PARV4 viral loads were arbitrarily expressed as log10 genome copies/150 000 cells using, respectively, 10 copies of PARV4 sequences and five copies of human beta-globin as the sensitivity of the PCR reactions [2,7].

PARV4-positive samples were directly sequenced in the region flanked by the screening primers using the Big Dye Terminator Kit (ABI PRISM, Foster City, California, USA). Sequence alignment was performed using the SeqScape software v2.5 (ABI) employing as reference the PARV ORF1 sequence AY622943 deposited in Gene Bank.

Sixteen patients (45.7%) were positive for PARV4 DNA sequences and all were infected with PARV4 genotype 1, as demonstrated by sequence analysis. A single nucleotide substitution in one of the amplified PCR products was detected, generating a synonymous mutation in the ORF1 gene.

PARV4 infection was most prevalent (11 of 16; 68.7%) in HCV-seropositive IDU. We also detected PARV4 in two out of 13 (15.4%) HIV-infected subjects without serological evidence of HCV infection and without a history of injection drug use (Table 1).

Table 1

Table 1

The PARV4 viral load never exceeded 3 log10 copies/150 000 cells except in one case (6.25%), whereas it was between 3 log10 and 10 copies/150 000 cells in the remaining 15 out of 16 patients (93.75%).

We failed to document the presence of PARV4 genotype 2 in our patients with AIDS. This may be explained by the fact that only five subjects in this series were born before 1956. In addition, they were all heterosexuals and were seronegative for HCV infection (i.e. less at risk for PARV4 infection).

Our preliminary survey suggests that PARV4 infection is readily detectable in Italian patients with AIDS, and is not restricted to HIV or HCV-infected IDU.

Finally, we documented low PARV4 viral loads in the BMA of our patients and an almost perfect sequence identity of the amplified viral products that confirm, as already suggested [4], ongoing low-level PARV4 replication in patients with AIDS and a relatively recent spread of this virus in Italian individuals with HIV.

Back to Top | Article Outline


1. Jones MS, Kapoor A, Lukashov VV, Simmonds P, Hecht F, Delwar E. New DNA viruses identified in patients with acute viral infection syndrome. J Virol 2005; 79:8230–8236.
2. Fryer JF, Kapoor A, Minor PD, Delwart E, Baylis SA. Novel parvovirus and related variant in human plasma. Emerg Infect Dis 2006; 12:151–154.
3. Fryer JF, Lucas SB, Padley D, Baylis SA. Parvoviruses PARV4/5 in hepatitis C virus infected patients. Emerg Infect Dis 2007; 13:175–176.
4. Manning A, Willey SJ, Bell JE, Simmonds P. Comparison of tissue distribution, persistence, and molecular epidemiology of parvovirus B19 and novel human parvoviruses PARV4 and human bocavirus. J Infect Dis 2007; 195:1345–1352.
5. Norja P, Hokynar K, Aaltonen LM, Chen R, Ranki A, Partio EK, et al. Bioportfolio: lifelong persistence of variant and prototypic erythrovirus DNA genomes in human tissue. Proc Natl Acad Sci U S A 2006; 103:7450–7453.
6. Antinori S, Calattini S, Longhi E, Bestetti G, Piolini R, Magni C, et al. Clinical utility of the polymerase chain reaction performed on peripheral blood and bone marrow in the diagnosis and monitoring of visceral leishmaniasis in HIV-infected and uninfected patients. A single center, 8-year Italian experience and review of the literature. Clin Infect Dis 2007; In press.
7. Pizzuto M, Piazza M, Senese D, Scalamogna C, Calattini S, Corsico L, et al. Role of PCR in the diagnosis and prognosis of visceral leishmaniasis in patients coinfected with human immunodeficiency virus type 1. J Clin Microbiol 2001; 39:357–361.
© 2007 Lippincott Williams & Wilkins, Inc.