[page 57↓]

3 . Results

Several different extraction- and PCR-methods have been employed in this study. Methods have been changed and were even substituted as soon as better methods were available. Different sample groups have been studied, which were collected (or received) at different times during the study period, depending mainly on external circumstances. Some sample goups (eg. Psammomys, desert rodents) have been proceeded with the purpose of evaluating the quality of the method. As a "byproduct", interesting and sometimes unexpected results were found, as for example a new host species. Therefore technological findings or improvements are inseparably connected with the specific results obtained from different sample groups. This chapter is mainly structured according to the methods used.

3.1. Efficiency of the tested DNA extraction methods:

A major objective has been to find the most efficient and practicable extraction method with particular regard to the dermal scrapings on filter paper. It was important to find a reliable method which could be adopted to a clinical laboratory for routine diagnosis. The results with the different extraction methods are summarized here and will be thoroughly discussed in chapter 4 (Discussion).

All tested extraction methods were successful. The classical phenol-chloroform method was applied on several series of dermal scrapings of patients (mostly the Hadassah patients at the beginning of the study). Differently preserved samples were occasionally received too (eg. Giemsa stained smears, paraffin embedded skin biopsies, formalin fixed jackal spleens). These samples were proceeded with the phenol-chloroform extraction, because this method had the greatest potential to remove inhibitory agents. The tested Giemsa stained smears (positive by microscopy) were PCR-positive (3 out of 3), as well as the paraffin embedded biopsies (4 out of 4). The formalin fixed tissue was not amplified with Leishmania specific primers, but yielded a weak signal with mammalian primers (performed by Carney Mattheson). This proved that formalin fixed tissue can be theoretically proceeded, but that inhibition has to be expected. The phenol-cloroform extraction is labor-intensive and difficult to handle in a non-research oriented laboratory. Due to the frequent re-opening of the 1.5 ml tubes, the risk of sample cross-contamination was highest with this method.

For these reasons other extraction methods were tested too: At first, experiments with crude extractions were performed, using only the first steps of the phenol-chloroform extraction (lysis and digestion, concluded by a short boiling step). Experiments with this crude method were [page 58↓] performed on dermal scrapings from CL lesions in the initial phase of the study, when only the kDNA primers Uni21/Lmj4 were employed. The success rate was sometimes high, (up to 4 positives out of 5 tested samples), and sometimes the results were poor (none, 1 or 2 positives in a series of 5-10 samples). Conditions were changed each time in order to finally introduce the best methodology (eg. amount of template, additives in the PCR, cycling temperatures). It was obvious that inhibition was the main cause for sub-optimal results: small amounts of template (1-2 m l lysate) had a far better success rate than larger amounts (5 m l), the addition of BSA reduced the inhibition by hemoglobin.

The guanidine extraction was tested as another potential alternative to the phenol-chloroform extraction. It also produced good results, on patient samples (mainly Tel Hashomer Hospital,Wadi Albethan) and on the dog peripheral blood spots (on filter paper). The guanidine method was easier to handle and also quicker than the phenol-chloroform extraction.

The Chelex extraction was only “discovered” much later as another possible extraction method. It has been employed on the rodent samples (ear scrapings from Psammomys and Gerbilli ). Since 28 out of 30 Psammomys samples (ear scrapings on filter paper) were strongly positive it can be stated that this method is highly efficient. These samples were characterized by a high tissue and a low blood content. The Chelex method was also employed on individual patients, but it has not been evaluated yet on a larger number of patient samples. One patient sample (dermal scraping) was incubated for 1 hour in Chelex (not overnight) and was amplified without problems. An experiment with one sample (infected mouse liver, spotted on filter paper) which was incubated for either 1, 2, 3 or 4 hours showed that a longer incubation was indeed more efficient (more PCR product, stronger bands on the gel). The Chelex method was even more simple than the previous crude method (lysis, digestion, boiling) so that it could replace the latter.

No differences have been noticed between the 3 extraction methods (phenol-chloroform, guanidine and Chelex) when compared with practically identical samples (infected mouse liver spotted homogenously on filter paper, squares of 0.5.cm 2 , 6 pieces for each method).

3.1.1 Suitability of different specimens for PCR diagnosis:

It was examined whether dermal scrapings (preserved on filter paper) would in fact be suitable for direct PCR diagnosis of CL. The sampling method and way of preservation proved to be efficient, as seen in several study groups (Hadassah patients, rodent ear scrapings, dog blood, Wadi Albethan inhabitants, Tel Hashomer patients). Filter paper samples (infected mouse liver spotted on filter paper) were stable at all three temperatures (20 ° C, 4 ° C, -20 ° C) for more than one year and were amplified equally well. Dermal scrapings on filter paper were examined [page 59↓] predominantly, but it was also relevant to test other types of samples. It was especially of interest to obtain results from early cultures in which promastigotes had been just detected. A crude preparation from Schneider´s medium (lysis in H 2 O and boiling for 5´minutes) was amplified well (no inhibition seen). Inhibition was observed when blood containing media were used (eg. semisolid media). Peripheral blood spots on filter paper (from dogs with CVL), Giemsa stained smears and paraffin embedded skin biopsies proved to be suitable too.

3.1.2 Experiments with crude samples:

Inhibition was observed as a regular phenomenon when crudely extracted samples were subjected to PCR. The experiments with additives for the prevention of PCR-inhibition showed the following: bovine serum albumin (BSA) clearly helped to amplify crude samples containing hemoglobin. The crude template (1 m l) was amplified well in the presence of 4 m l BSA (10 mg/ml), less amplified with 2 m l of BSA, very faintly amplified with 1 m l BSA and not amplified without BSA (results not shown). The experiment with 2.5 % DMSO and 1% formamid also showed a positive effect: When the additives were used the reaction tolerated 1, 2.5 and 5 m l of template. Without the additives the reaction with 5 m l of crude template was inhibited. Results are shown in Figure 9.

Figure 9: Experiment with additives against inhibition:


[page 60↓]

3.2. Amplification of whole minicircle DNA with primers Uni21/Lmj4:

3.2.1. Results with DNA purified from cultured reference strains:

It was found that the size of the PCR product was consistent within each species. Whole minicircles of kDNA were amplified, the PCR products therefore reflected the size of the minicircles in each species, ranging from 680 bp ( L.major ), 800 bp ( L.donovani complex), 820 bp ( L.tropica ) to 850 bp ( L.aethiopica ). L.major could easily be differentiated from L.tropica and the L.donovani complex by the size of the PCR products, which had been one important aim of the study. L.tropica and the L.donovani complex could not be distinguished certainly since both produced similar sizes of bands. The minor bands were not always seen. The appearance of these bands depended on the PCR conditions (eg. the Taq-polymerase). Even though some of them seemed to be related to the species they were not consistent enough to be used for species specific identification. The New World species ( L.mexicana, L.braziliensis, L.guyanensis, L.panamensis, L.amazonensis and L.d.chagasi ) could be clearly distinguished from the Old World species either by not amplifying at all or yielding much smaller products. Only L.d.chagasi, as a species belonging to the L.donovani complex, was amplified (820 bp). Species of Leishmania that are not pathogenic to humans were amplified: L.gerbilli, L.turanica and L.killicki yielded a product of 800 bp. A 680 bp product was obtained from L.arabica . Results with reference strains are shown in Figure 10:

Figure 10: PCR with primers Uni21/Lmj4 on different Kinetoplastidae (20 ng of DNA)


[page 61↓]

Kinetoplastidae other than Leishmania were not amplified with primers Uni21/Lmj4 (Trypanosoma cruzi, Trypanosoma lewisii, Crithidia fasciculata, and Phytomonas davidi). The DNA of Leptomonas seymouri was amplified but the product was considerably smaller (500 bp). Primers Uni21/Lmj4 are specific for the Old World species of Leishmania, with the only exception of L.d.chagasi (identical to L.d.infantum in the Old World).

The sensitivity of the PCR assay with primers Uni21/Lmj4 was examined with a serial dilution of L.major DNA. If PCR products were visualized with Gel star staining, maximum sensitivity was found to be at 25 fg of target DNA (less than one parasite). With regular ethidium bromide staining, dilutions of less than 1 pg were almost undetectable. The sensitivity study is shown in Figure 11:

Figure 11: Comparison of sensitivity with Gel Star staining and ethidium bromide

3.2.2. Results from lysed cultured isolates with primers Uni21/Lmj4:

PCR proved to be reliable whenever cultures of promastigotes cultivated in Schneider´s medium were used as lysates. If the cultures were bloody because of the medium (NNN and semisolid medium contain rabbit blood), the PCR reaction sometimes failed. Most patients had CL, acquired in Israel, where L.major and L.tropica were expected to be the causative agents. The smaller product (680 bp) was identified as L.major, whereas the larger product (about 800 bp) was either amplified from L.tropica or from the L.donovani complex. Based on the patient history (CL or VL, origin of infection) and on additional tests (eg. EF), the species was determined as being either L.tropica or of the L.donovani complex. Culture 14 (Table 5, LRC- [page 62↓] 747) was derived from a sandfly trapped in a cave near Kfar Adumim (Photo 13) in the Judean Desert (trapped by Alon Warburg, Kuvin Center). It yielded a product of about 800 bp, which practically confirmed the suspected species of L.tropica . (The focus was already known to be endemic with L.tropica , since cutaneous infections by L.d.infantum have never been reported in the country). L.tropica was later confirmed by IFA, EF and PPIP-PCR (by colleagues at the Kuvin Center). Interestingly, L.major has also been identified from Kfar Adumim residents in 3 patients (Tables 5 and 6). L.major transmission in this area is possible too, since the Jordan V alley with a high L.major incidence is very close and endemicity in the settlement itself can not be excluded.

Culture 19 (LRC-744) was derived from a child with VL in Costa Rica. (The strain was brought to Israel by Lionel Schnur, Kuvin Center). It yielded a PCR product of 800 bp. L.d.chagasi was diagnosed since no other American Leishmania species would have been amplified. Table 5 and Figure 12 show the results from lysed cultures.

Table 5: PCR results with Uni21/Lmj4 from lysed cultures:

No.

Sample

Smear

PCR

Diagnosis

Origin

1

P1092

+

680 bp

L.major

Negev

2

P1111

++

680 bp

L.major

Kfar Adumim

3

P1120

+

680 bp

L.major

Jericho

4

P1130

+

680 bp

L.major

Jericho

5

P1132

-

680 bp

L.major

Jordan Valley

6

P1135

+

680 bp

L.major

Jericho

7

P1136

-

680 bp

L.major

Jericho

8

P1125

-

680 bp

L.major

Jordan Valley? Negev?

9

P1161

+

680 bp

L.major

Israel

10

P1165

-

680 bp

L.major

India?/Israel?

11

P2001

n.a.

680 bp

L.major

Israel

12

LRC-757

n.a.

820 bp

L.tropica

Kfar Adumim, sandfly

13

LRC-768

n.a.

820 bp

L.tropica

Israel

14

LRC-747

n.a.

820 bp

L.tropica

Kfar Adumim, sandfly

15

LRC-765

n.a.

680 bp

L.major

India

16

LRC-767

n.a.

680 bp

L.major

India

17

LRC-773

n.a.

680 bp

L.major

Israel

18

LRC-691

n.a.

680 bp

L.major

Israel

19

LRC-744

n.a.

800 bp

L.d.chagasi

Costa Rica

20

LRC-762

n.a.

680 bp

L.major

Uzbekistan

The LRC-number is given to newly collected strains of the Leishmania R eference C enter, Jerusalem. The corresponding smears were not available (n.a.) . P = patient of the Hadassah Hospital, Jerusalem.


[page 63↓]

Figure 12: PCR with primers Uni21/Lmj4 on lysed cultures:

3.2.3. Results with primers Uni21/Lmj4 with dermal scrapings on filter paper:

This part of the work served to establish methods, which required experimental changes of conditions almost in every extraction series and PCR. This resulted in different success rates. Therefore the results of this part of the work are presented as a collection of positive results without a statistical evaluation.

Skin scrapings of patient lesions (spotted onto filter paper) proved to be useful for the diagnosis of CL, in patients as well as in reservoir animals. Positive results helped in species diagnosis, negative results could not exclude infection . It turned out that a low success rate was often due to inhibition. Series in which 2 m l of template were used had a better success rate than 5 m l eg.. The DNA lost its quality quickly (resuspension in H 2 O), limiting the possibility of re-amplification of the relatively large PCR-product (680-850 bp). Out of this reason it was difficult to compare different PCR conditions with the same samples. The best possible DNA quality was necessary to obtain results, therefore freezing was not considered in this case (with smaller fragments no problem, see later). Out of filter paper samples from 100 suspected CL patients 24 were found positive with primers Uni21/Lmj4, either with the regular phenol-chloroform extraction or with the crude method ( proteinase-K digestion, lysis and boiling). The first extractions yielded excellent results (also with the crude method!) as shown in Figures 13 and 14, whereas later extractions sometimes failed completely, with not even one single positive in 10 samples (mostly due to inhibition). The PCR from filter paper was usually successful when the corresponding smears contained a high number of amastigotes. When the smears (Giemsa stained) contained only single amastigotes the PCR with primers Uni21/Lmj4 (using the corresponding filter paper [page 64↓] sample) the method was often not sensitive enough . Most patients were diagnosed with L.major (23 out of 24), one patient was diagnosed with L.tropica (Figure 13 lane 10, Table No. 6, patient 6). However, 6 patients with CL were diagnosed by PCR in whom microscopy and culturing had failed. Table No. 6 represents the results with whole amplified minicircles (Uni21 / Lmj4) on dermal scrapings (preserved on filter paper).

Table 6: Results with primers Uni21/Lmj4 from skin scrapings preserved on filter paper:

No

Patient

Smear

Culture

PCR

Diagnosis

Origin of infection

1

1075

-

-

680 bp

L.major

Ethiopia/ Israel ?

2

1077

+

-

680 bp

L.major

Sinai, Arava

3

1078

-

-

680 bp

L.major

Israel

4

1083

-

-

680 bp

L.major

Dead Sea

5

1087

+

-

680 bp

L.major

Dead Sea

6

1070

++

-

820 bp

L.tropica

Hemdat

7

1088

++

-

680 bp

L.major

Negev

8

1095

+

-

680 bp

L.major

Jordan Valley

9

1100

++

-

680 bp

L.major

Holon

10

1101

++

-

680 bp

L.major

Holon

11

999

++

-

680 bp

L.major

South America/Israel ?

12

1102

+

-

680 bp

L.major

South America/Israel ?

13

1108

+

-

680 bp

L.major

Jordan Valley?

14

1124

-

-

680 bp

L.major

Kfar Adumim

15

1111

++

+

680 bp

L.major

Kfar Adumim

16

1113

+

+

680 bp

L.major

Qziot, Negev

17

1135

+

+

680 bp

L.major

Jericho

18

1150

++

-

680 bp

L.major

Jericho

19

1152

+

-

680 bp

L.major

Jordan Valley

20

1143

-

-

680 bp

L.major

Baghdad/Israel ?

21

1164

-

-

680 bp

L.major

Sinai

22

1181

+

-

680 bp

L.major

Qziot, Negev

23

1193

++

-

680 bp

L.major

Sinai

24

LRC-762

?

+

680 bp

L.major

Uzbekistan

25

3 Psammomys

+

n.a.

680 bp

L.major

Qziot, Negev

26

Mouse liver

+

+

800 bp

L.donovani

Infection in the laboratory

n. a.- not available. For better understanding: The origin of infection relates to the patient history. It contains the information obtained from the patient before the diagnosis was made.

Figure 13: PCR with primers Uni21/Lmj4 on crudely extracted filter paper samples:


[page 65↓]

Figure 14: PCR with primers Uni21/Lmj4 on filter paper extractions

3.2.4. Selected patient cases diagnosed with kinetoplast primers Uni21/Lmj4

Results obtained from dermal scrapings on filter paper

1 . One patient suffered from a severe and prolonged infection of the nose (photo 1) ; he had a second lesion on his hand. He was a young man of 20 years coming from Hemdat , an Israeli settlement, located in a hilly area close to the Jordan Valley . The direct analysis from the skin scrapings on filter paper (1 of each lesion) revealed a band of 820 bp ( Figure 13 lane 10, Table No. 6, patient 6) from one lesion, and two bands from the other lesion (680 bp and 820 bp, the latter band being weaker, not shown). The PCR performed on the culture some weeks later yielded a band of 680 bp . These contradictive results will be discussed in chapter 4.

2 . A married couple suffered from multiple lesions (12, respectively 9) distributed all over the body (Photo 5) . They came from Holon, an urban center close to Tel Aviv and claimed they had not travelled either inside or outside the country for at least one year. It was the first case of CL in this part of the country and the origin of infection could not be identified. There was only a vague hypothesis that a sandfly might have been trapped in the car of a neighbour, who was travelling frequently to the Jordan Valley. PCR diagnosis succeeded from tissue scrapings in the couple from Holon (patients 9 and 10 in Table 6). L.major was diagnosed (680 bp product ). The smears were positive but cultures failed to grow.

3. A visiting researcher from Uzbekistan had been vaccinated with a viable L.major strain. A lesion had developed at the vaccination site, which had no tendency to heal even after several months and was forming also satellite lesions (Photo 4). Even though the cause of the infection was clear, a skin scraping was collected, to test if the PCR would in fact confirm the infection of [page 66↓] Leishmania major . The PCR confirmed L.major (patient 24 in Table 6; lane 13, Figure 12-PCR also from the culture).

4. Ear scrapings from Psammomys (No. 25 i n Table 6, Figure 16) were examined and a product of 680 bp was found in all 3 individuals, thus confirming the expected species, L.major .

5. The positive extraction controls (infected mouse liver, No. 26 i n Table 6) always yielded a 800 bp product, as expected for the L.donovani complex.

6. Patients 11 and 12 (Table 6) were diagnosed with L.major, and New World Leishmania species could be excluded as causative agent. The infections were therefore contracted in Israel and not during the journey in South America.

3.3. Genus specific amplification with kDNA primers 13A/13B:

All Kinetoplastidae ( Leishmania, Trypanosoma, Crithidia ) tested in the PCR with primers 13A/13B yielded a product of 120 bp. The amplification of the positive extraction control indicated successful DNA extraction. The sensitivity with a serial dilution of purified DNA was at 10 fg (data not shown) with ethidium bromide staining (Gel Star staining not needed). This confirmed the sensitivity found by Rodgers et al ., (1990) who had introduced the primers and had found a sensitivity corresponding to far less than one parasite. The negative extraction controls (negative filter paper, human blood on filter paper) of the first few extraction series were contaminated. Only after potential sources of contamination were eradicated, by following strictly the precautions mentioned in chapter 2 ( Material and M ethods), the problem was controlled. At the time the contamination problems were controlled with the more sensitive, genus specific PCR the patient samples (which had been mainly used to establish the diagnosis with primers Uni21/Lmj4) were used up. Primers 13A/13B were evaluated with new sample groups, which were received later (eg. animal samples). For species specific identification some of the PCR-positive samples were submitted to PCR with primers Uni21/Lmj4. This succeeded with 3 out of the 4 tested Psammomys samples, and L.major was confirmed (Figure 16).

3.4. Studies on reservoir animals:

1. Desert rodents: (Photo 18)

The animal results are summarized in Table 7. As expected, the PCR proved to be more sensitive than microscopy on the Psammomys ear scrapings: Every microscopically positive smear was also positive by PCR. The genus specific PCR with primers 13A/13B detected Leishmania infections in 93.3 % (28 out of 30) of the tested P.obesus samples. Out of the 28 PCR-positive samples 20 were positive by microscopy (71.4%). Eight out of 10 smear-negative samples were thus positive by PCR. (Microscopy was performed by Gideon Wasserberg, Ben Gurion [page 67↓] University, Beer Sheva). Some of the smears were double checked in our laboratory, the microscopical results were in concordance between the two lab oratories. The amplifications were highly efficient, as most samples showed intens ive bands (Figure 15). The negative Psammomys control ( laboratory animal from the Diabetes U nit , Hadassah Hospital ) was always negative (as expected). As mentioned already L.major was confirmed with primers Uni21/Lmj4 (Figure 16) and later also with ITS-1 amplification and RFLP-analysis (Figure 17). In the latter experiment non human pathogenic Leishmania species (reference strains) were amplified and restricted for comparison.

According to PCR, 65% (11 out of 17) ear scrapings from Gerbillus dasy a rus w ere positive (Figure 15) . The PCR products were seen as faint bands. A new host species was identified (discussed in chapter 4). After positive PCR results the smears were examined once more. One single amastigote was finally detected in one of the smears and approved by three examinators , thus confirming the findings by PCR. The faint bands of the PCR products correlated with the scarc i ty of amastigotes in the smears. This explains also why the Gerbillus dasyarus sample failed to amplify in the ITS-1 PCR (Figure 17). The larger band (Figure 15, lane 15) may be a non-specific amplification seen sometimes also in amplifications from dog samples. A reliable negative control of the same animal species was not available which could have helped to clarify this non-specific amplification.

Figure 15: Psammomys obesus and Gerbillus dasyarus
PCR with primers 13A/13B, Chelex extraxted ear scrapings on filter paper


[page 68↓]

Figure 17: PCR with ITS-1 primers and restriction with BsuRI ( HAE III):

2. Canids:

The 4 tested dogs (Photo 20) were positive by PCR with primers 13A/13B (results not shown). Interestingly, a repeated extraction and PCR performed by a collegue showed the same intensities of bands on the gel. This indicated a positive correlation between antigen load and the quantity of amplified product. The 21 jackal ears were all negative. The extraction and the PCR had been successful as shown by the positive extraction- and positive PCR-controls. The 14 extracted jackal spleens in formalin were PCR-negative too. In order to test, if the negative results were due to inhibition or poor DNA quality mammalian primers were tested on 3 of the jackal samples (by Carney Matheson, guest scientist at the Kuvin Center, University of Queensland, Australia). Faint signals were seen, meaning that amplification was principally possible from formalin fixed samples, but that this preservation method was probably not [page 69↓] suitable for sensitive diagnosis. It remains unclear, if any of the jackals were Leishmania -positive.

3. Hyrax:

All hyrax samples (parched and fresh tissue samples) were PCR-negative. A hyrax is shown in Photo 19.

Table 7: Results from animal samples:

species

Specimen

No. of samples

PCR +

smear +

Psammomys

ear scrapings

30

28

20

Gerbillus dasyarus

ear scrapings

17

11

1

Meriones crassus

ear scrapings

4

0

0

dogs with CVL

EDTA blood

4

4

n.a.

Jackal

ear biopsies, frozen

20

0

n.a.

Jackal

spleens in formalin

14

0

n.a.

Hyrax

nose, skin, bone

6

0

n.a.

Hyrax

ear biopsy, frozen

1

0

n.a.

mouse

liver tissue

positive control

always positive

positive

hamster

foot pad aspirate

1

positive

n.a.

CVL =canine visceral leishmaniasis; n.a. =not available

3.5. Specific d etection of L eishmania braziliensis with kDNA primers MP3H/MPL1:
Results obtained from dermal scrapings preserved on filter paper:

It was important to establish PCR diagnosis also for Leishmania infections contracted in the New World. Since especially L.braziliensis infections are feared due to the risk of MCL the most important aim was to identify L.braziliensis . The PCR with primers MP3H/MPL1 was successfully employed on the reference strains as well as on the patient samples. All tested Leishmania species of the L.braziliensis complex ( L.braziliensis, L.guyanensis, L.panamanensis ) were amplified and yielded a product of 70 bp (in Figure 18 the band appears to be slightly larger, the band was usually exactly at 72 bp of the marker). L.mexicana and the Leishmania species of the Old World did not amplify, as expected. The method proved to be highly specific with the only exception that L.amazonensis ( L.mexicana complex) was sometimes amplified with a very faint signal , which was not expected. A carry-over contamination between the New World reference DNA strains can not be excluded. The DNA had been handled by several different persons previously. Contamination could have occurred by pipetting the different species into PCR tubes for one experiment , keeping in mind that this PCR method is capable of detecting DNA of far less than one parasite (Lopez et al., 1993). No other L.amazonensis strain was available. The sensitivity was known to be at least as high as with the genus specific primers 13A/13B, according to the literature (Belli et al ., 1998) . Serial dilutions with purified DNA give [page 70↓] generally only theoretical sensitivities, which are not necessarily reproduced with clinical samples. Therefore it was preferred to directly evaluate the sensitivity on the clinical samples, in comparison with the genus specific primers (13A/13B). The sensitivity was 6/6 (100%) of the patients who had visited Tuiji in Bolivia and who were suspected to be infected with L.braziliensis . The genus specific primers gave positive results in all 11 patients who were examined (4 patients infected in the Old World, 1 in Guatemala). The follow-up study (repeated PCR with the genus specific and the L.braziliensis specific primers directly after and months after treatment) indicated that the L.braziliensis specific primers MP3H/MPL1 were even slightly more sensitive than the genus specific primers 13A/13B (eg. patients 1, 2 in Table 8).

Patients of the Tel Hashomer Hospital:

Skin scrapings and cultures were taken from 11 patients who were admitted to the Tel Hashomer Hospital ( Tel Aviv ) for systemic treatment with Pentostam. Species specific diagnosis especially of the L.braziliensis complex was required since 7 patients had contracted CL in the New World (Photos 15-17). Six patients had visited Tuiji, an attractive tourist site in the Bolivian rain forest, 1 patient had been trave l ling in Guatemala. The other 4 patients had been infected in the Old World, 1 patient in Turkmenistan, 1 patient in Afula (Galilee), 1 in Korazim, a village on the northern slope of the lake of Galilee and another patient had been infected in the Negev . The patient from Afula was suffering from a severe infection of the nose (Photo 2) , the patient from Korazim suffered from a lesion, which involved one eyelid (Photo 3). Both cases were of special interest, since L.tropica was expected to be the causative agent, and both places had not been identified as being L.tropica foci previously . The diagnosis of CL had been made in all 11 patients beforehand in the hospital, by microscopy of skin biopsies . All 11 patients yielded positive results with the genus-specific primers 13A/13B. The results are specified together with the results of primers MP3H/MPL1 in T able 8.

The patient who was infected in Guatemala (patient 12) was negative with the L.braziliensis specific primers MP3H/MPL1 and positive with the genus specific primers 13A/13B, which was suggestive of an infection with L.mexicana . Patients 4, 5, 8 and 9 were infected in the Old World. All were positive with the genus specific primers and negative with the L.braziliensis specific primers, thus confirming the specificity of the primers.

Repeated PCR, directly after treatment with pentostam and also several months after completing the therapy, was still positive in most cases, though weaker. In some cases only one of the two PCR methods reproduced signals after repeated sampling (after treatment). Specific results are documented in Table No 8 and Figures 18-20. Cultures were positive in patients 1 (from 2 [page 71↓] different lesions), 4, 7 and 9 (collected before treatment). This was the first time that L.braziliensis has been diagnosed in Israel.

Figure 19: Genus specific PCR with primers 13A/13B (120 bp)

Patients 1, 2, 4 before treatment, patient 3 after systemic treatment with pentostam. Dermal scrapings on filter paper, guanidine extracted. Lane 1 neg. FP; Lane 2 neg. blood; lane 3 patient 1a (BO); lane 4 patient 1b (BO), (taken from another lesion); lane 5 patient 2 (BO); lane 6 patient 3 (BO); lane 7 patient 4 (Afula, IL); lane 8 patient 1201 (IL); lane 9 patient 1203 (IL); lane 10 patient 1204 (IL); lane 11 patient 1187 (IL), lane 12 negative PCR-control, lane 13 L.major; lane 14 L.tropica.

Figure 20: Genus specific PCR with primers 13A/13B:

[page 72↓] Patients 1-4 after treatment with pentostam. Lane 1 negative filter paper; lane 2 negative blood, lane 3 patient 1 (BO); lane 4 patient 2 (BO); lane 5 patient 3 (BO); lane 6 patient 4 (Afula, IL); lane 7 and 8 patient 5a, 5b (Korazim, IL), (2 paraffin embedded skin biopsies taken from two lesions); lane 9 and 10 paraffin embedded skin biopsies from Tchechia, patient infected in Jordan; lane 11 Hyrax ear; lane 12 sandfly (West Bank); lane 13 L.major; lane 14 negative PCR control; lane 15 L.mexicana; lane 16 L.braziliensis

Table 8: Results of patients from the Tel Hashomer Hospital:

Patient Nr.

1

2

3*

4**

5**

6

Origin of infection

Tuiji, Bolivia

Tuiji, Bolivia

Tuiji, Bolivia

Afula, IL

Korazim, IL

Tuiji, Bolivia

1 st Sampling

8.5.00

8.5.00

8.5.00

8.5.00

1.6.00

11.7.00

1 st PCR genus

+/+

+

+

+

+

+/+

1 st PCR L.braz.

+/+

+

+

-

-

+/+

Culture

+/+

n.d.

n.d.

+

-

-

Identification

L.braziliensis

L.braziliensis

L.braziliensis

L.tropica?

L.tropica?

L.braziliensis

2 nd sampling

1.6.00

1.6.00

1.6.00

1.6.00

 

5.9.00

2 nd PCR genus

-

-

+ weak

+

+

+

2 nd PCR L.braz.

+ weak

+ weak

+

-

-

++

3 rd sampling

5.9.00

relapse?

10.10.00

26.12.00 relapse?

11.7.00 relapse?

 

10.10.00

3 rd PCR genus

-

-

 

+

 

+

3 rd PCR L.braz.

+ weak

-

 

-

 

+

4 th sampling

relapse?

 

9.1.01

 

 

19.11.00

4 th PCR genus

-/-/+ weak

 

 

 

 

+/- weak

4 th PCR L.braz.

+/+/+

 

-

 

 

-/-

5 th sampling

19.11.00

2 nd treatment

 

 

 

 

23.01.01

5 th PCR genus

+/+ weak

 

 

 

 

 

5 th PCR L.braz.

-/-

 

 

 

 

+

Patient Nr.

7

8 ***

9 ***

10

11

Origin of infection

Tuiji, Bolivia

Negev, IL

Turkmenistan

Tuiji, Bolivia

Guatemala

1 st Sampling

23.10.00

12.11.00

12.11.00

21.11.00

10.10.00

1 st PCR genus

n.d.

+/+

+/+/-

-/+/+ weak

+

1 st PCR L.braz.

+

?

-/-/-

+/+/+ weak

-

Culture

+

-

+

 

 

Identification

L.braziliensis

L.major ?

L.major ?

L.braziliensis

L.mexicana?

2 nd sampling

19.10.00

12.11.00

 

24.12.00

 

2 nd PCR genus

+?

+/+

 

+

 

2 nd PCR L.braz.

-

 

 

 

 

3 rd sampling

19.11.00

 

 

march 2001

 

3 rd PCR genus

+ weak

 

 

 

 

3 rd PCR L.braz.

+ weak

 

 

+/+

 

4 th sampling

9.1.01

 

 

 

 

4 th PCR genus

 

 

 

 

 

4 th PCR L.braz.

+

 

 

 

 

5 th sampling

march /01

 

 

 

 

5 th PCR genus

 

 

 

 

 

5 th PCR L.braz.

+

 

 

 

 

L.braz. =means L.braziliensis. *Patient 3 had already had treatment before the first visit. ? indicates that the species was suspected, according to the patient history and the results obtained so far. Species were confirmed later with ITS-1 PCR: **Patient 4 and 5 later were confirmed with L.tropica,***Patients 8 and 9 were L.major, patient 9 with kinetoplast primers Uni21/Lmj4 from the early culture.

Figures 19 and 20 represent diagnostic PCR experiments (from dermal scrapings on filter paper) at two different times. Several patients are identical in both figures, with the difference that Figure 19 shows results before treatment and Figure 20 shows results after the 3 weaks course of treatment with pentostam: Bands in lanes 3 and 4 are derived from two different lesions of patient 1. Figure 20 shows that no amplification was seen after treatment in patient 1 (lane 3, only one lesion re-examined). The same was found in patient 2 (lane 5 Figure 19, lane 4 Figure 20). Patient 3 (lane 6 in Figure 19) was examined only after treatment and then re-examined later (lane 5 Figure 20). This patient had just completed the treatment when the sampling was performed for the first time. As seen in this patient and also in patient 4 (lane 7 in Figure 19, lane 6 in Figure 20) DNA was amplified also after the treatment, though weaker. This will be discussed in chapter 4. The other patients are not identical in the two figures, because it was naturally the case that new cases were added to the series whenever diagnosis was required in new patients visiting the outpatient clinic for travellers.

3.6. Amplification of the ITS-1 region and RFLP analysis:

All tested species (purified DNA of WHO reference strains) were amplified. The products had a size of about 300 bp, with minor differences seen between some of the species, eg. L.braziliensis products appeared to be slightly smaller (not shown). The sensitivity was up to 1 pg when tested with purified DNA. Out of 36 previous ly positive extractions (thawed for reuse in this PCR) 25 were successfully amplified and submitted to restriction with the enzyme BsuRI. The tested samples showed the restriction patterns as it was expected (species specific diagnosis had been made previously in most cases) . Out of 5 patients from the Hadassah-Hospital 4 were infected with L.major and 1 patient with L.tropica . The recently discovered endemic sites Wadi Albethan (Photos 10, 11) and Korazim were confirmed as being L.tropica foci, as well as the infection seen in patient 4 from Afula (Galilee) (Table 8, Figure 21 lane 2). Some L.tropica amplifications have been comparatively weak, the smaller expected band at 57 bp is therefore not seen (in lanes 3 and 5, compared to lane 10, reference strain). Therefore these L.tropica results appear to be similar to the L.d.infantum reference (lane 11). The patients who were infected in Tuiji (Bolivia) were once more confirmed with an infection due to L.braziliensis . Table 9 shows the expected sizes of restriction fragments of each species and the results obtained from dermal scrapings on filter paper. Figure 21 shows one example of ITS-1 amplification and BsuRI- digestion.


[page 74↓]

Figure 21: PCR with ITS-1 primers and restriction with BsuRI:

Table 9: BsuRI restriction fragments of ITS-1 amplificates:

Species

Fragments after Restriction with BsuRI (bp)

Results

L.major

203

132

P1167 (IL)

P1202 (IL)

P1175 (IL)

P1168 (IL)

3 P.obesus , Qziot (IL)

L.tropica

185

57

53

24

Wadi Albethan No.19

P 4 Afula (IL)

P1201 (IL)

P5 Korazim (IL)

L.donovani complex

184

72

54

hamster liver on fp

infected with L.donovani (2x)

L.aethiopica

200

57

54

23

 

L.braziliensis

156

143

P1 (BOL) (2x)

P2 (BOL)

P7 (BOL)

P10 a/b (BOL)

L.guyanensis

156

137

 

L..panamensis

156

139

 

L.mexicana

186

88

59

 

L..amazonensis

186

142

 

L.turanica

203

54

52

24

 

3.7. A new focus of L.tropica in Wadi Albethan, West Bank:

One major purpose of the project was the study of the endemic sites in the West Bank, since most of them were not sufficiently understood and some of them were not documented yet. A new L.tropica focus has been identified in the country, in an Arab village named Wadi Albethan. The area of Wadi Albethan is mountainous, declining towards the Jordan Valley (Photos 10 and 11). Wadis (dry riverbeds) are connecting with the Jordan Valley. A brook is running through the village. In the center of the village settlement is dense, in the periphery houses are more isolated, surrounded by meadows with various kinds of trees and bushes (predominantly almond trees). Animals (predominantly goats) are kept in open sheds. The inhabitants claimed the infections only occurred for about one year. The characteristic lesions were not known before to the local population. Nevertheless, one young man had been suffering from leishmaniasis recidivans (lupoid leishmaniasis) for 9 years (Photo 8) . He presented with a scarring lesion on his cheek, with active eruptions on the border of the lesion, and with one separate lesion on the nose. Another youth had an old scar (Photo 9). Interestingly, the inhabitants reported that hyraxes (Photo 19) had become abundant over the past two years. Due to the altitude of this area a focus of L.tropica was suspected. Several visits to this village were undertaken. In some houses several family members suffered from cutaneous lesions, some of them had multiple lesions (Photo 6). The lesions had appeared between 1 to 4 months before, in one person the lesion had evolved earlier (8 months before). Adults and children were equally affected. The severity of the lesions ranged from small papules to extensive and deep ulcers. Ulcers located on the nose tended to be severe. Some of the ulcerative lesions were obviously super-infected with bacteria (Photo 7). Some of the patients had received treatment with either liquid nitrogen or intralesional pentostam. Some infected residents had not seen a doctor and were left untreated. A visit to a doctor and subsequent treatment appeared to depend on the income of the family. Some families could not afford medical care. The information presented here is based on observations made during sporadic visits; a systematical epidemiological survey has not been carried out yet.


[page 76↓]

The preservation of skin scrapings on filter paper proved to be a suitable preservation method for PCR-diagnosis in this field study. The PCR with primers 13A/13B succeeded at first in 12 out of 23 guanidine extracted samples. This provided a very good result, considering that the methods were applied for the first time in a real field study in an unknown focus. A second extraction and re-amplification using only 2 m l template instead of 10 m l showed that all patients were positive , indicating inhibition by some components in the DNA-extract (the repeat was performed by Kefaya Azmi, Al-Quds-University). Primers Uni21/Lmj4 failed to directly prove the species, but later L.tropica was confirmed in one tested sample by ITS-1 amplification (Figure 20, Table 9) and BsuRI- digestion. L.tropica was also confirmed by other methods based on the successfully cultured strains (EF, Lionel Schnur and PPIP-PCR, Carol Eisenberger). The microscopical examination of the smears of 23 patients from Wadi Albethan revealed amastigotes in 9 smears, which is a sensitivity of 39%. Four smears had to be classified as negative, even though the microscopical picture was highly suggestive of a leishmanial infection: structures in the size of amastigotes were seen in groups either intracellular in macrophages or extracellular, but no nuclei and kinetoplasts could be identified. The following table shows epidemiological data and diagnostic results of the examined residents of Wadi Albethan.

Table 10: Results from patients from Wadi Albethan:

No.

Sex

Age

Duration of lesions

in months

Number of lesions

PCR 13A/13B

Culture

Smear

Therapy?

1

f

~12

2

2

+

n.d.

++

 

2

f

~11

1

1

+

n.d.

++

 

3

f

13

1

1

+

-

-

1 P

4

f

10

1

1

+

-

-

1 P

5

m

25

8

2

+

-

-

no

6

m

48

3

6

+

-

-

N

7

f

12

1

1

+

-

-

N

8

f

35

2

4

+

n.d.

+

N

9

m

13

4

3

+

-

-

3 P

10

m

50

3

2

+

-

-

4 P

11

m

3

4

1

+

n.d.

+

1 P

12

m

3

3 weeks

1

+

n.d.

-

no

13

m

23

2

4

+

n.d.

++

no

14

m

43

4

3

+

n.d.

-

4 P

15

m

27

1

1

+

-

+

no

16

m

15

1

1

+

-

++

?

17

m

44

4

5

+

+

+

2 P

18

m

19

3

1

+

n.d

-

2 N

19

f

18

1

1

+

n.d

-

no

20

m

4

3

3

+

n.d

-

3 P

21

m

1

3

1

+

n.d

+

?

22

m

4

1

3

+

n.d

-

no

23

f

23

1

7

+

+

-

1 P

f =female; m =male; n.d. =not done; contam. =culture contaminated with bacteria; P =intralesional pentostam; N =liquid nitrogen. The number refers to the number of doses received; no =no treatment received

3.8. The aims of the study have been achieved:

  1. Direct PCR diagnosis has been established in the laboratory and is ready to be used in clinical laboratories. The tested sampling procedure (dermal scrapings on filter paper) proved to be a highly suitable method which can be recommended for future applications. Diagnosis of CVL(VL) is available too, using peripheral blood spots on filter paper. Many observations have been made, related to preservation, extraction and amplification. This may help to select methods according to different purposes. Extraction methods have been simplified with regard to the application in clinical laboratories (eg. Chelex extraction).
  2. 2. The differentiation of L.major and L.tropica has been achieved. Primers Uni21/Lmj4 were able to make the distinction but the diagnosis from dermal scrapings did not always succeed. With the recent introduction of the ITS-1 primers a sensitive system was established for the differentiation of nearly all species. The latter is also able to distinguish L.d.infantum infections, the third endemic species in the country.
  3. PCR diagnosis directly from lesions has been established also for the L.braziliensis complex specifically, using kinetoplast primers MP3H/MPL1. This can help to decide if patients with New World leishmaniasis need to be hospitalized for 3 weeks or not. Species-specific diagnosis of the New World Leishmania species had not been available in the country before. The introduction of the ITS-1 method improved the quality of diagnosis, because it could also positively identify the L.mexicana complex (including the distinction of L.mexicana and L.amazonensis ).
  4. It has been shown that the methods employed (DNA extraction and PCR) were suitable for screening of reservoir animals. PCR was far more sensitive than microscopy, which clearly justified the method for diagnosis and for epidemiological studies.

In addition, some important results have been obtained: A new host species has been discovered ( Gerbillus dasyarus ) and a new focus of L.tropica has been identified (Wadi Albethan, West Bank).

[page 78↓] Photography: L.tropica infections which required systemic treatment with pentostam

Photo 1: L.tropica , Hemdat, IL

Photo 2: L.tropica Afula, Galilee, IL

Photo 3: L.tropica Korazim, Galilee

[page 79↓]Cutaneous leishmaniasis by L.major

Photo 4: Lesion with satellite, complication of vaccination
with a L.major strain in scientist from Uzbekistan

Photo 5: Mutiple lesions by L.major in Israeli patient from Holon, near Tel Aviv


[page 80↓]

Wadi Albethan, West Bank (Samaria), outbreak of L.tropica:

Photo 6: multiple lesions

Photo 7: Superinfected lesion

Photo 8: Leishmaniasis recidivans (lupoid leishmaniasis) for 9 years, L.tropica

[page 81↓]Wadi Albethan, West Bank (Samaria), outbreak of L.tropica:

Photo 9: CL scar

Photo 10: Wadi Albethan, view to the east, towards the Jordan Valley (behind mountain range)

Photo 11: Area of high incidence in Wadi Albethan

[page 82↓]Endemic sites of Cutaneous and Visceral Leishmaniasis


Photo 12: Jericho, Jordan Valley, L.major


Photo 13: Kfar Adumim, Judean Desert, L.tropica


Photo 14: Jenin district, West Bank, L.d.infantum

[page 83↓]L.braziliensis infections in Israelis who were infected in the Bolivian rainforest

Photo 15: L.braziliensis, before treatment with pentostam

Photo 16: L.braziliensis, before treatment with pentostam

Photo 17: L.braziliensis, after treatment with pentostam

[page 84↓]Reservoir animals

Photo 18: Psammomys obesus, (Fat sand rat) desert rodent, reservoir of L.major

Photo 19: Hyrax, En Gedi, suspected reservoir of L.tropica

Photo 20: Dog with CVL, Nataf, infected with L.d.infantum


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 3.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated:
08.12.2003