No subject

m VL could be anthroponotic ! The prevalence in humans is extremely low, an=
d the parasitic loads in the blood are generally very low. By contrast, the=
 prevalence in dogs is very high (>80%) and the disease is 'viscero-cutaneo=
us', so that parasites can be found everywhere, obviously in blood and the =
(often diffusely) infected skin, but also in healthy skin ! Now, if these a=
uthors can explain to me how parasites from human stay in human and do not =
infect dogs, or vice-versa?....

By contrast, there is a huge bias in this study, which is that 27 out of th=
e 28 human strains analysed in this study are from cutaneous cases (CL) ! L=
. infantum CL is often caused by different zymodemes (so-called 'variant' b=
y Rioux and colleagues) than the hugely prevalent MON-1. This is corroborat=
ed here by the inclusion of the CL reference strains (# 2 and 3, zymodemes =
MON-24 and MON-29) among the 'human' subpopulation. These two zymodemes cau=
se essentially CL, except in HIV+ patients where they may cause VL (Montpel=
lier's data, but also several publications, e.g. Kallel et al. Acta Trop 20=
08).
It would have been interesting to carry out isoenzyme analysis on these str=
ains, but unfortunately very few labs in the world still perform isoenzyme =
analysis (perhaps our colleagues from Rome or Barcelona? in Montpellier, we=
 still have a bit of know-how, but have now stopped this activity). The aut=
hors used a k26-PCR to identify so-called 'MON-1' strains, and found only 7=
 'non-MON-1' strains, again among human cases.
Finally, the authors examined the geographical distribution of the strains =
and showed a further sub-structuring, correlating with geographical situati=
on, among human strains but not canine strains. My opinion is that, due to =
human activities and to high transmission, canine strains 'travel' a lot mo=
re and interbreed a lot more than human strains, and are thus often  monomo=
rphic in dog populations.
Please note that the term "cutaneous" is not addressed in their discussion.
In the end, these results may suggest that CL strains would have a distinct=
 epidemiological cycle from that of VL (dog) strains. However, this is not =
either corroborated by previous studies: (i) dogs can also host CL zymodeme=
s; (ii) the sandfly vectors appear to be the same for both types of strains=
.

Kind regards
Patrick

Pr Patrick Bastien
Head
Department of Parasitology-Mycology
National Reference Centre for Leishmanioses
Molecular Biology 'Pole' of the National Reference Centre for Toxoplasmosis
Deputy Director of the Research Unit "MiVEGEC" (CNRS 5290 - IRD 224 - Unive=
rsity of Montpellier)
FACULT=C9 DE M=C9DECINE & C.H.U. de Montpellier
France