OUP user menu

Bartonella quintana in head louse nits

Emmanouil Angelakis, Jean-Marc Rolain, Didier Raoult, Philippe Brouqui
DOI: http://dx.doi.org/10.1111/j.1574-695X.2011.00804.x 244-246 First published online: 1 July 2011


The body louse is the principal vector of Bartonella quintana, the causative organism of trench fever, but B. quintana DNA has also been detected in adult head lice. Because there are no characteristics that distinguish the body louse from the head louse, we decided to test head louse nits collected from a homeless man for the presence of B. quintana DNA. All of the sampled nits tested positive by real-time PCR, and intergenic spacer region (ITS) gene sequences shared 100% similarity to the corresponding ITS fragment of the genome of B. quintana. The role of the head louse in the maintenance and transmission of B. quintana remains to be determined.

  • Bartonella quintana
  • head louse
  • homeless
  • nits

Head and body lice have been recognized as human parasites for thousands of years (Light et al., 2008). Head lice live on the scalp and lay eggs at the base of hair shafts, whereas body lice live in the clothes on the surface of the body and lay their eggs on clothing. The body louse is medically relevant as this organism can transmit life-threatening bacteria to humans. Among these bacteria are Rickettsia prowazekii, which causes epidemic typhus, Bartonella quintana, the agent of trench fever, and Borrelia recurrentis, the louse-borne relapsing fever spirochete (Brouqui, 2011). With the exception of a few phenotypic characteristics, such as the length of the tibia on the second pair of legs, studies of the morphological characters and primary endosymbionts of head and body lice suggest that these two types of lice are conspecific (Sasaki-Fukatsu et al., 2006). Moreover, genetic analysis has not been able to show any differences between these two subspecies; all molecular data available at this time show no reciprocal monophyly between head and body lice and indicate that no known species concept would recognize these louse morphotypes as separate species (Light et al., 2008). A study by Takano-Lee (2003) demonstrated that head lice can be reared successfully in vitro through a complete life cycle. It has also been proposed that body louse heterozygosity would be retained and would transform into the head louse phenotype if the organism were relocated to the head or reared under head louse conditions (Levene & Dobzhansky, 1959). In both the homeless of developed countries and in Nepalese children, dual infestation by both body and head lice was found in up to 50% of the cases examined, and under these circumstances, it is impossible to differentiate between the head and the body lice (Brouqui, 2011). Finally, with the current state of knowledge, head lice are defined by their location on the head and by the attachment of their eggs (called nits) to the base of the hair shaft, while body lice are defined by their location in the folds of clothes. Head lice infestation occurs most often in school children, but has also been reported in the homeless and other impoverished populations (Sasaki et al., 2006). Although Fournier and colleagues successfully identified B. quintana and R. prowazekii in body lice collected from people living in poor conditions, such as the homeless and refugees (Badiaga et al., 2008), their earlier attempts to identify the presence of B. quintana and R. prowazekii in 143 head lice collected from school children across eight countries were not successful (Fournier et al., 2002). The first evidence that B. quintana might infect head lice was shown by Sasaki (2006), who detected the bacterial DNA in both head and body lice from two heavily infested Nepalese children who were living on the streets or in slum areas. Bonilla et al. (2009) reported the detection of B. quintana DNA in the head lice of a homeless individual from San Francisco, CA, without any known concurrent body louse infestation. Because the body and the head louse are phenotypically indistinguishable, we decided to test head louse nits for B. quintana DNA.

Since 2000, we have collaborated with homeless shelters in Marseille, France, to investigate arthropod-borne diseases in this socially displaced and deprived population (Badiaga et al., 2008; Brouqui, 2011). The study was approved by our ethical committee under No. 10-005. We regularly visited the shelters and looked for body lice. During this survey, we met a homeless male who presented with a massive head lice infestation. While vagabond's leukomelanoderma, a hallmark of past body louse infestation, was noted, no presence of evolving scratching lesions was recorded. A conscientious clinical examination revealed neither adult head nor adult body lice on the removed clothes. Unfortunately, a blood sample was not collected at the time of sampling head louse nits. After obtaining the patient's consent, we shaved the patient and found only head louse nits. Some nits were found about 3–3.5 cm from the hair follicle. Most nits were full of larvae, and some had begun hatching (Fig. 1). DNA was extracted from three sets of pooled nits using a QIAamp tissue kit (Qiagen, Hilden, Germany). The DNA was used as a template in a real-time PCR assay targeting a portion of the Bartonella 16S–23S intergenic spacer region (ITS), as described previously (Fournier et al., 2002), and two specific B. quintana genes (fabF3 and yopP) that encode for 3-oxoacyl-[acyl-carrier-protein] synthase and a hypothetical intracellular effector, respectively. The reactions were performed with the following primer sequences: (Bqui05300F/FabF3/5′-GCT-GGC-CTT-GCT-CTT-GAT-GA-3′, Bqui05300R/FabF3/5′-GCT-ACT-CTG-CGT-GCC-TTG-GA-3′, Bqui11580F/yopP/5′-TAA-ACC-TCG-GGG-GAA-GCA-GA-3′, Bqui11580R/yopP/5′-TTT-CGT-CCT-CAA-CCC-CAT-CA-3′) and probes (Bqui05300P/FabF3/6FAM-TGC-AGC-AGG-TGG-AGG-AGA-ACG-TG-TAMRA, Bqui11580P/yopP/6FAM-CGT-TGC-CGA-CAA-GAC-GTC-CTT-GC-TAMRA). The specificity of these encoding genes was verified in silico as well as on a panel of 14 different Bartonella species (data not shown). All three sets were positive by real-time PCR. Amplicons were then purified using the QIAquick Spin PCR purification kit (Qiagen, Courtaboeuf, France) and sequenced on an ABI 3100 automated sequencer (Perking Elmer, Courtaboeuf, France) using the dRhodamine Terminator cycle-sequencing ready reaction kit (PE Applied Biosystems, Les Ulis, France), according to the manufacturer's instructions. Sequences obtained after sequencing of the ITS gene shared 100% similarity to the corresponding ITS fragment of the genome of B. quintana (Toulouse strain) (GenBank accession number NC_005955). To cultivate B. quintana, a fourth set of pooled nits was triturated in brain–heart medium before inoculation onto Columbia 5% sheep blood agar plates (BioMerieux, Marcy l'Etoile, France). Plates were placed in polyethylene bags and incubated at 37 °C in 5% CO2 (Genbag CO2 system; BioMerieux) (La Scola & Raoult, 1999). This assay was repeated independently three times, and the agar plates were examined weekly. After 1 month, no evidence of growth was observed.

Figure 1

Nits collected from the hair of a homeless individual. Living larvae within the nit sheathes and one larvae hatching (white arrow) are shown. The photograph was taken using a Nikon light microscope at a magnification of × 20.

As the speed of hair growth is roughly 1.25 cm month−1, we can estimate that the head louse infestation in the examined homeless individual occurred at least 3 months back; however, because larvae were apparent through the nit's sheaths, the infestation appeared to be ongoing (Fig. 1). To avoid the potential of misinterpretation of the PCR data due to laboratory contamination, three different genes were assessed, and all of them were successfully detected. The source of the Bartonella DNA could have been the larvae of the louse, the nit's sheath or the cement deposit from the accessory glands of an infected female head louse. DNA contamination of the hair by means of body louse feces containing B. quintana is unlikely, as no adult body lice were found on the homeless individual's clothes, and no scratching lesions were noted, indicating that the individual was not currently infested with body lice.

We were not able to cultivate B. quintana from nits or larvae. This result suggests that either the number of living microorganisms in the tested pool was too small to be isolated by culturing or that only DNA, but no living organisms, was present in the nits.

Interestingly, all attempts to detect B. quintana in the head lice of school children have failed (Fournier et al., 2002). Bartonella quintana DNA has been detected exclusively in head lice collected from impoverished populations such as the homeless or Nepalese children living in slums or on the streets, who are usually infested by both head and body lice (Sasaki et al., 2006; Bonilla et al., 2009). In alcoholic homeless individuals, B quintana bacteremia can be prolonged for weeks (Foucault et al., 2006), and under these circumstances, one could hypothesize that head lice might be contaminated by blood that is infected with B. quintana. Nonetheless, the role of the head louse in the maintenance and transmission of B. quintana remains to be determined.


This article has been edited in English by the American Journal Expert under No. WN29KQY3.


This study was supported by CNRS UMR 6236. None of the authors have a conflict of interest relevant to this study.


  • Editor: Ake Forsberg


View Abstract