AIDS RESEARCH AND HUMAN RETROVIRUSES Volume 20, Number 1, 2004, pp. 105–109 ©Mary Ann Liebert, Inc.

Sequence Note

Protease in HIV-1 Non-B Strains Infecting Drug-Naive Villagers of Cameroon

FRANK A. J. KONINGS, 1 PING ZHONG, 2,3 MOISE AGWARA, 9 LUCY AGYINGI, 4 LEOPOLD ZEKENG, 5 JACQUELINE M. ACHKAR, 6 LEONARD EWANE, 5 DR. SAA,7 EMMANUEL AFANE ZE, 7 THOMPSON KINGE, 7 and PHILLIPE N. NYAMBI 2,8

ABSTRACT

To describe the presence of protease inhibitor (PI) resistance-associated mutations and subtype distribution in drug-naive villagers of six provinces of Cameroon, we sequenced the protease (PR) (297 bp) of 128 . Secondary PI resistance-associated mutations were identified at five sites: L10I/V (16%), K20R (8%), M36I (98%), L63P (13%), and V77I (6%). No primary in the PR was identified. Of the 128 speci- mens analyzed, subtypes A (11%), C(2%), D (6%), F2 (3%), G (6%), H (0.8%), J (6%), and CRF02_AG (60%) were identified. The mutations identified were not characteristic to any particular subtype. The ab-

sence of primary mutations, in addition to the few secondary mutations, gives good perspectives for PI treat- ment interventions in these rural areas.

NSPITEOFTHEEMERGENCEOFVARIOUSDRUGRESISTANCE -AS- little or no influence on resistance. Naturally occurring PI re- ISOCIATEDMUTATIONS in the viral protease (PR) gene, pro- sistance-associated secondary mutations have different preva- tease inhibitors (PIs) contribute substantially to the control of lence patterns for non-B subtypes and subtype B viruses. The the AIDS epidemic in Europe and North America. However, of most common natural polymorphisms in a panel of 187 non-B the estimated 42 million people infected with human immuno- strains from drug-naive individuals worldwide were reported to deficiency type 1 (HIV-1) worldwide, the majority (29.4 be M36I (83%), L63P (17%), L10I/V (13%), K20R (10%), and million) live in sub-Saharan Africa, 1 the region with the broad- V77I (2%).4 est HIV-1 diversity. Since PIs were designed and tested for sub- In Cameroon, the majority of the population lives in rural type B viruses only, they could be less effective in countries, areas. While our recent reports reveal the HIV-1 in like Cameroon, in which a broad genetic diversity exists. 2 the rural villages is characterized by a broad genetic diversity, 5,6 PI resistance is usually established during treatment by an information on the natural polymorphisms that could be rele- initial primary mutation followed by gradual selection of sec- vant to antiretroviral drugs, such as PIs, is sparse. Here we pro- ondary mutations. 3 Primary mutations, which are located in the vide detailed insight into the naturally occurring mutations that conserved regions of the PR gene, cause PI drug resistance by could lead to PI resistance in HIV-1-infected villagers in six themselves, while secondary mutations further reduce drug sus- provinces of Cameroon. This information is needed to estab- ceptibility by increasing the replicative fitness of the virus, with lish a baseline for monitoring resistance to PIs and in the

Departments of 1Microbiology and 2Pathology, New York University School of Medicine, New York, New York 10010. 3Shanghai Municipal Center for Control and Prevention, Peoples Republic of China. 4Faculty of Science, University of Dschang, Dschang, Cameroon. 5Laboratoire de Santé Hygiè ne Mobile, Yaoundé, Cameroon. 6Department of Medicine, New York University School of Medicine, New York, New York 10010. 7Ministry of Public , Yaoundé, Cameroon. 8Research Center for AIDS and HIV Infection, VA Medical Center, New York, New York 10010. 9Faculty of Science, University of Yaoundé 1, Yaoundé, Cameroon.

105 106 KONINGS ET AL. successful introduction of antiretrovirals in the rural regions of PR region by removal of the 5 9 and 39 ends. Sequences are Cameroon and neighboring countries. available from GenBank with accession numbers AY359683– This work was part of a study aimed at determining HIV in- AY359810. fection in rural villages in Cameroon. 6,7 From a total of 300 Phylogenetic analysis of the aligned sequences was per- HIV-1-positive plasma samples, 128 were randomly selected formed by the neighbor-joining method of TREECON (Treecon for analysis of the PR gene. These samples were obtained from for Windows, version 1.3b). 9 Distance calculation was per- individuals living in villages in six provinces of Cameroon, in- formed by Kimura’s two-parameter method. The statistical ro- cluding Center ( n 5 26), East ( n 5 13), South ( n 5 30), South- bustness of the neighbor-joining tree and reliability of the west (n 5 20), West ( n 5 28), and Northwest ( n 5 11) between branching patterns were confirmed by bootstrapping (1000 2000 and 2002. Of the 128 specimens analyzed, 56 were from replicates). The aligned DNA sequences were translated to males and 72 from females. Their ages ranged from 20 to 70 amino acids by GeneRunner (GeneRunner 3.05 for Windows, years for males (mean 39 years) and from 17 to 66 years for http://www.generunner.com). females (mean 31 years). None of the individuals in this study Previous studies have indicated that the 297-bp PR gene al- had previously received any kind of antiretroviral treatment. lows subtyping of nonrecombinant HIV-1 group M subtypes A, Viral RNA was extracted from plasma as previously de- B, C, D, F, G, H, J, and K. 10,11 Our results are in accordance scribed by Boom et al.8 One-tube reverse transcriptase poly- with these observations and show that most of the strains (95%) merase chain reaction (RT-PCR) (Access RT-PCR System; could be subtyped, while only 5% were unclassifiable (U). Promega, Madison, WI) was performed for amplification of 297 Phylogenetic analysis revealed seven group M subtypes (A, bp of the PR region, corresponding to HXB2 positions C, D, F2, G, H, J), and CRF02_AG, while no subtype B was 2258–2552. Two forward primers NYUPOL6 (5 9-AGGGAAG- identified among the sequences analyzed. The HIV-1 group M GCCAGGGAATTT-3 9, HXB2 location 2114– 2132) and subtype distribution in the six provinces studied is shown in NYUPOL7 (59-GGGAATTTTCTTCAGAGCAG-3 9, HXB2 Table 1. CRF02_AG was the predominant subtype in all the location 2125– 2144) in combination with the reverse primer provinces and accounted for 60% of all the isolates in this study. NYUPOL8 (59-TCTTCTGTCAATGGCCATTGT-3 9, HXB2 HIV-1 subtypes A and G and CRF02_AG viruses were identi- location 2615–2635) were used for RT-PCR. The two forward fied in all of the six provinces, except for the Northwest primers (NYUPOL6 and NYUPOL7) together in the same mix, province where no subtype G was found. As reported earlier, 7 in combination with a single reverse primer (NYUPOL8), al- the South and West provinces both had the broadest HIV-1 di- lowed the amplification of more PR sequences. For nested versity with six different group M subtypes and unclassifiable PCR the primers were NYUPOL9 (5 9-TCCTTAACTTCCCT- strains.

CAAATCACT-3 9, HXB2 location 2241– 2264) and NYU- The 297-bp PR sequences were translated to the corre- POL10 (59-CTGGCACGGTTTCAATAGGACT-3 9, HXB2 lo- sponding 99 amino acids and analyzed for primary and sec- cation 2556– 2577). For both RT-PCR and nested PCR the ondary mutations that are associated with in vivo PI resistance annealing temperature was 48°C. as reported for subtype B viruses. 12 We found five secondary PCR products were separated by agarose gel electrophore- amino acid substitutions in the PR, namely L10I/ V, K20R, sis, gel extracted using the QIAquick spin gel extraction kit (QI- M36I, L63P, and V77I. The prevalence of amino acid substi- AGEN, Valencia, CA) and directly DNA sequenced using an tutions for the 128 sequences examined was 16% (L10I/V), 8% automated DNA sequencer (373XL; Applied Biosystems, (K20R), 98% (M36I), 13% (L63P), and 6% (V77I). Except for Forster City, CA). The sequences were aligned with HIV-1 PR one sequence, all the sequences examined had at least one sec- reference sequences of various subtypes from the HIV data- ondary mutation. No specific amino acid mutation was re- bases at http:/ /www.hiv-web.lanl.gov. Multiple alignments stricted to sequences from a particular province. However, we were performed automatically by CLUSTAL X with minor noted that the sequences from the Center province were char- manual adjustments and the sequence was cut to the 297-bp acterized by a high frequency of L10I/V (27%) and L63P (23%)

TABLE 1. HIV-1 PROTEASE SUBTYPE DISTRIBUTIONIN SIX PROVINCES OF CAMEROON

Number of HIV-1 group M subtype Number Province tested A AGa C D F2 G H J Ua

Center 26 2% 18% — 2% — 1% 1.8% — 2 East 13 1% 9% — — — 1% — 2% — South 30 3% 14% — 4% 2% 3% — 1% 3 Southwest 20 4% 10% — 2% — 2% — 2% — West 28 3% 16% 3% — 2% 1% — 2% 1 Northwest 11 1% 10% — — — — — — —

Total 128 14% 77% 3% 8% 4% 8% 1.8% 7% 6 Percentb 11% 60% 2% 6% 3% 6% 0.8% 6% 5%

aAG, CRF02_AG; U, unclassifiable. bPercentage of total number of samples analyzed for each subtype. HIV-1 PROTEASE MUTATIONS IN CAMEROON 107

(data not shown). The most common mutation among the five polymorphisms that would evade drug treatment. Analysis of codons was M36I, which was present in 125 (98%) of the 128 the PI resistance-associat ed mutations revealed that primary sequences examined, while V77I (6%) was rare. mutations were absent, although five secondary mutations were Next, we examined these non-B subtype sequences to iden- identified in all the 128 sequences examined. Since secondary tify PI resistance-associat ed mutations and to determine mutations have little or no effect on resistance in the absence whether certain mutations predominated among specific non-B of primary mutations, the presence of these mutations should subtypes. The of these different mutations of the re- be considered when starting PI treatment. Analysis of multiple spective subtypes is shown in Table 2. We found that no mu- mutations further revealed that the majority of the sequences tation predominated any particular subtype. The only two sub- had harbored only single mutations (67%) and that dual muta- types that harbored all five mutations in some isolates were tions could be found in only 27% of the sequences. The subtypes A and F2. The three subtype C sequences had only CRF02_AG viruses accounted for the largest group of viruses the M36I mutation. Though the majority of the sequences ex- identified in this study. However, it is striking that almost all amined belonged to CRF02_AG (60%), this group of sequences (99%) viruses harbor the M36I mutation and only two addi- revealed only three different mutations, including M36I (99%), tional secondary mutations (L10I/V and L63P) were identified L10I/V (8%), and L63P (8%), and lacked K20R and V77I mu- in 8% of the sequences. The large number of M36I could in- tations. Subtype G viruses had the same three mutations as the dicate that this is a natural polymorphism in non-B subtypes CRF02_AG viruses (M36I, L10I/V, and L63P). Both subtypes analyzed in this study. It remains to be confirmed whether in- D and J viruses harbored four mutations (L10I/V, K20R, M36I, dividuals harboring isolates containing multiple secondary mu- and V77I), though only one (14%) of the seven subtype J and tations may be at greater of virological failure during PI one (13%) of the eight subtype D viruses harbored all four mu- therapy, as has been hypothesized.14,15 However, the results of tations. The single subtype H sequence identified contained our study in which the majority of the sequences contained only only two mutations, K20R and M36I. single mutations suggest that the PR of these viruses have Additionally, we investigated the sequences for the presence not evolved sufficiently to allow for evasion of PI treatment of multiple secondary mutations. While only one sequence had and as such these drugs could be as effective as they are on no secondary mutations, 86 (67%) of the 128 sequences har- subtype B viruses. However, to get a complete picture of re- bored a single amino acid mutation, followed by 34 (27%) se- sistance to other antiretroviral drugs, further studies are war- quences with two mutations. In two (2%) sequences, four mu- ranted to examine naturally existing resistance mutations to tations with identified and only one sequence had accumulated drugs directed at other regions of the viruses, like the reverse five mutations. Of the AG sequences, 84% (65/77) had single transcriptase.

mutations and 13% (10/ 77) had dual mutations (data not A total of five secondary amino acid mutation sites (codons shown). 10, 20, 36, 63, and 77) were found in the sequences examined. Finally, we examined all 128 sequences for PR mutations Of these five sites, mutations at codons 10, 20, and 36 occur in that have not been reported to be involved in drug resistance up to 2– 6% of drug-naive persons infected with subtype B and found that the following mutations were present in more viruses.16,17 Mutations at these same sites occurred in a greater than 40% of the sequences: I13V (81%), K14R (53%), K20I proportion (8–98%) of the non-B subtype sequences examined (67%), E35D (41%), R41K (89%), P63L (73%), H69K (85%), and also occur at higher rates in non-B subtype isolates rang- and L89M (85%) (Table 2). All eight mutations were found to- ing from 10 to 83% as reported by others. 14,18,19 With the ex- gether in sequences of subtypes A, G, J, and CRF02_AG, while ception of the M36I mutation that predominates non-B subtypes only three of them were found in subtypes C, D, and H. Each (95–97%) as observed in this study and by others, 11,20 other mutation was present in more than 80% of the CRF02_AG se- mutations occurred at codon 10 (16%) and 20 (8%) with a sim- quences, except for E35D, which was present in 30% of the se- ilar prevalence as observed in subtype B isolates [13% quences. (L10I/V), 10% (K20R)]. 4,18,19 The frequency of mutations at This study has examined the presence of PI resistance-asso- codons 63 and 77 in our study was 13 and 6%, respectively. ciated mutations in PR sequences of drug-naive HIV-1-infected Codon 63 is the most polymorphic PR position, but only L63P villagers living in six provinces of Cameroon. Our analysis re- is implicated in PI resistance. 17 While studies of non-B sub- vealed a broad HIV-1 diversity with seven group M subtypes types indicate that about 45% of isolates in untreated patients including subtype A, C, D, F2, G, H, J, and CRF02_AG. carry the L63P mutation, we observed that this mutation was CRF02_AG viruses and group M subtype A viruses were found present in only 13% of the sequences analyzed. Taken together, in all the provinces and caused the majority of all these patterns of mutations suggest some additional differences with 60 and 11%, respectively. These data support our previ- in the natural polymorphisms of these viruses from the rural ous studies in rural villages that revealed a broad env and gag villages to viruses from other regions of the world and thus subtype diversity and dominance of CRF02_AG infections. 7 should be considered when designing new PIs for use world- With the hypothesis that a broad genetic diversity within a pop- wide. ulation suggests a longstanding presence of the viruses in that Of interest was the identification of several amino acid mu- population, the results of our studies suggest that these viruses tations that have not been associated with PI resistance but were must have been in these populations for many decades as pre- present in more than 40% of the sequences. These mutations viously proposed. 13 and the proportion of sequences carrying them included I13V In the context of this broad genetic diversity, we wanted to (81%), K14R (53%), K20I (67%), E35D (41%), R41K (89%), study whether genes relevant to drug treatment have evolved P63L (73%), H69K (85%), and L89M (85%). All these muta- and whether these viruses have acquired characteristic natural tions are located in the variable regions of the PR. Whether any 108 KONINGS ET AL. ) ) ) ) ) ) 0 0 0 0 ) ) 3 2 0 0 0 0 0 5 M 9 9 1 1 1 1 5 8 9 ( ( ( — ( ( — ( ( (

8 3 1 3 4 8 7 3 9 L 1 7 0 1 ) ) ) 0 ) 0 ) 0 ) ) 0 7 0 5 0 0 5 K 1 9 1 7 1 5 8 9 ( ( ( — — ( ( ( ( 6

4 5 3 6 1 7 3 9 H 1 7 0 E S 1 A E T ) O ) ) 0 ) ) ) ) ) ) R L 1 8 0 5 0 8 4 7 3 P 3 7 8 1 2 5 3 1 6 7

( ( ( ( ( ( — ( ( ( 6 1

- P 0 8 3 2 2 3 1 4 3 V 1 6 9 I d H e t ) ) ) ) N a I i 0 ) 0 0 0 ) ) ) c 0 5 0 0 0 4 3 9 S K o 1 9 1 1 1 1 8 8 N 1 s ( ( — ( ( ( ( ( ( O s 4 I a 4 3 8 4 8 1 1 5 4 T R - 1 7 1 A e 1 T c R U E n a G t M A ) ) s i L ) ) 0 0 ) ) ) ) D s L D 0 0 0 0 8 4 3 1 E I e 5 5 3 1 1 8 1 3 4 T r ( ( — ( ( ( — ( ( ( V 3 n A

I o E E 7 3 8 4 7 1 2 2 C 2 5 V N O I ) S A S % ( N A

) - - s ) 0 ) ) ) G E I n ) 7 0 4 7 7 U C 0 o 7 9 1 1 1 6 R i N 2 ( ( — — — ( — ( ( ( t

A D a K 1 5 8 1 1 6 t T S 7 8 u G I N S m I E

T f R C ) ) ) ) o N

E ) 1 0 9 3 R r F O 7 8 5 2 5 4 e ( ( — — — ( — ( — ( N N 1

I b 1 2 4 2 8 K D m S 6 6 E u N P A N Y T D )

B E ) ) 0 ) ) ) T U 3 7 0 9 3 1 V A S 9 9 1 2 8 8 I 3 ( ( — — — ( ( ( ( C

1 M O I 3 5 8 1 2 5 4 S 1 7 0 P S 1 U A O - R E C ) ) ) G I ) 3 5 1 ) N 7 T 7 1 2 7 6 A 7 ( — — ( ( — — ( — ( N T

V E S 1 1 1 5 8 I R S E E F d F R I ) ) ) ) ) ) e t 1 ) 8 0 5 7 3 R D P a 2 8 3 5 2 1 1 i O 3 ( ( — ( ( ( — — ( ( F c T

6 I O o 3 6 3 2 2 1 7 L B s I 1 S s H E a - N C I e N c ) ) ) ) ) ) E E n 0 ) 0 0 0 0 ) 0 ) S U I a 0 9 0 0 0 0 6 0 8 A Q t 6 1 9 1 1 1 1 8 1 9 s E E ( ( ( ( ( ( ( ( ( i 3 T

S s 4 6 3 8 4 8 1 6 6 5 O M e r R 1 7 2

1 P I P Y

R y ) ) ) ) r A 1 5 4 3 ) a R D 2 7 1 3 8 . d 0 N ( — — — ( — ( ( ( e n 2

O l o 3 3 1 1 2 0 C b K c 1 a E e i S f S i s s ) ) ) ) ) ) ) . a V 4 ) 8 5 5 4 0 6 2 l / I 1 8 3 7 2 1 5 1 c E ( ( — ( ( ( — ( ( ( 0 n

L 1 u 2 6 3 3 2 1 3 0 B

L , 2 A U T

; r G e d A b 4 7 3 8 4 8 1 7 6 8 e t _ 1 7 2 m s 2 1 e u 0 t N F R C

, e p G l y t a a A t b a G a o 2 u S A A C D F G H J U T HIV-1 PROTEASE MUTATIONS IN CAMEROON 109 of these mutations will become relevant to escape from new PIs ware package for the construction and drawing of evolutionary trees remains to be seen. It is important that the influence of these for the Microsoft Windows environment. Comput Appl Biosci mutations on PI resistance is investigated and considered when 1994;10:569–570. designing new PIs that could be used in these regions where 10.Fonjungo PN, Mpoudi EN, Torimiro JN, et al.: Presence of diverse human immunodeficiency virus type 1 viral variants in Cameroon. the viruses prevail. AIDS Res Hum Retroviruses 2002;16:1319–1324. In conclusion, this study shows that a broad HIV-1 diversity 11.Fonjungo PN, Mpoudi EN, Torimiro JN, et al.: Human immuno- circulates in the rural villages of Cameroon. These HIV-1-in- deficiency virus type 1 group M protease in Cameroon: Genetic fected drug-naive villagers harbor viruses with single or dual diversity and protease inhibitor mutational features. J Clin Micro- secondary PI resistance-associat ed mutations. The absence of biol 2002;40:837–845. primary mutations and the few secondary mutations portends 12.Birk M and Sonnerborg A: Variations in HIV-1 pol gene associ- well for PI treatment interventions in these rural populations as ated with reduced sensitivity to antiretroviral drugs in treatment- they have been carried out for subtype B infections. naive patients. AIDS 1998;12:2369–2375. 13.Korber B, Muldoon M, Theiler J, et al.: Timing the ancestor of the HIV-1 pandemic strains. Science 2002;88:1789–1796. 14.Perez EE, Rose SL, Peyser B, et al.: Human immunodeficiency ACKNOWLEDGMENTS virus type 1 protease genotype predicts immune and viral responses to combination therapy with protease inhibitors (PIs) in PI-naïve We thank the several doctors, nurses, and social workers in patients. J Infect Dis 2001;183:579–588. the different regions of Cameroon for their assistance, and the 15.Perno CF, Cozzi-Lepri A, Balotta C, et al.: Secondary mutations individual volunteers who made this study possible. We also in the protease region of human immunodeficiency virus and vi- thank the Minister of Public Health, Mr. Urbain Olanguena rologic failure in drug-naïve patients treated with protease inhibi- Awono for his support in these studies. tor-based therapy. J Infect Dis 2001;184:983–991. Financial support: National Institute of and Infec- 16.Hertogs K, Bloor S, Kemp SD, et al.: Phenotypic and genotypic tious (grant AI 47053); National Institute of Health analysis of clinic HIV-1 isolates reveals extensive protease inhib- itor cross-resistance: A survey of over 6000 samples. AIDS (grant TW01254); Fogarty International Center at the NIH 2000;14:1203–1210. (grant TW01409); Funds from the Department of Veterans Af- 17.Kantor R, Machekano R, Gonzales MJ, Dupnik K, Shapiro JM, fairs (Merit Review Award). and Shafer RW: Human immunodeficiency virus reverse tran- scriptase and protease sequence database: An expanded model in- tegrating natural language text and sequence analysis. Nucleic REFERENCES Acids Res 2001;29:296–299.

18.Cornelissen M, van den Burg R, Zorgdrager F, Lukashov V, and 1.UNAIDS/ WHO: AIDS Epidemic Update. Sub-Saharan Africa, Goudsmit J: Pol gene diversity of five human immunodeficiency 2002, pp. 16–18. virus type 1 subtypes: Evidence for naturally occurring mutations 2.Korber B and Hoelsher M: HIV-1 subtypes: Implications for epi- that contribute to drug resistance, limited recombination patterns, demiology, pathogenicity, vaccines and diagnostics. AIDS 1997; and common ancestry for subtypes B and D. J Virol 1997;71: 11:17–36. 6348–6385. 3.Shafer RW: Genotypic testing for human immunodeficiency virus 19.Gonzales MJ, Machekano RN, and Shafer RW: Human immuno- type 1 drug resistance. Clin Microbiol Rev 2002;15:247–277. deficiency virus type 1 reverse-transcriptase and protease subtypes: 4.Pieniazek D, Rayfield M, Hu DJ, et al.: Protease sequences from Classification, amino acid mutation patterns, and prevalence in a HIV-1 group M subtypes A–H reveal distinct amino acid mutation Northern California clinic-based population. J Infect Dis patterns associated with protease resistance in protease inhibitor- 2001;184:998–1006. naive individuals worldwide. AIDS 2000;14:1489–1495. 20.Vergne L, Peeters M, Mpoudi-Ngole E, et al.: Genetic diversity of 5.Zhong P, Bundas S, Koning F, et al. Genetic and biological prop- protease and reverse transcriptase sequences in non-subtype-B hu- erties of HIV-1 isolates prevalent in villages of the cameroon equa- man immunodeficiency virus type 1 strains: Evidence of many mi- torial rain forest and grass fields: Further evidence of broad HIV nor drug resistance mutations in treatment-naive patients. J Clin type 1 genetic diversity. AIDS Res. Human Retroviruses 2003;19: Microbiol 2000;38:3919–3925. 1167–1178. 6.Nyambi P, Zekeng L, Kenfack H, et al.: HIV infection in rural vil- Address reprint requests to: lages of Cameroon. J Acquir Immune Defic Syndr 2002;31: Phillipe N. Nyambi 506–513. Department of Pathology 7.Zhong P, Burda S, Urbanski M, et al.: HIV type 1 group M clades New York University School of Medicine infecting subjects from rural villages in equatorial rainforests of c/o V.A. Medical Center Cameroon. J Acquir Immune Defic Syndr 2002;31:495–505. 423 E. 23rd Street, Room 18124N 8.Boom R, Sol CJA, Salimans MMM, Jansen CL, Wertheim-van Dillen PME, and van der Noordaa J: Rapid and simple method for New York, New York 10010 purification of nucleic acids. J Clin Microbiol 1990;28:495– 503. 9.van de Peer Y and de Wachter R: Treecon for Windows: A soft- E-mail: [email protected]