Collection records for the lice used in this study are as follows: Coloceras sp. ex zebra dove (Geopelia striata), Honolulu, Hawaii, USA by Sarah Bush, Aug. 6th 2001; Ibidoecus bisignatus (Nitzch) ex white-faced ibis (Plegadis chihi), Arcadia Pa., Louisiana, USA by Kevin P. Johnson, Sept. 3rd 2000; Anaticola crassicornis (Scopoli) ex chilöe widgeon (Anas sibilatrix), Rio Negro, Argentina by Kevin P. Johnson, Apr. 15th 2001 (host voucher KGM257); Philopterus sp. ex field sparrow (Spizella pusilla), McClain Co., Illinois, USA by Kevin P. Johnson, May 16th 2001; Quadraceps sp. ex bronze-winged courser (Cursorius chalcopterus), Buipe, Ghana, by Kevin P. Johnson, Mar. 10th 2003; Damaliniasika ex sika deer (Cervus nippon yesoensis), Nakajima, Lake Toyako, Hokkaido, Japan, March 7th 2002, Kazunori Yoshizawa. (voucher ID: KY397).
For Ibidoecus the 8 possible long PCR combinations of forward and reverse primers aiming to link the three partial gene sequences for cox1, rrnL and rrnS yielded two successful amplicons, rrnL to cox1 (IBID3/IBID5, approx. 8000 bp in size) and cox1 to rrnL (IBID6/IBID4, approx. 8000 bp). No PCRs originating in the preliminary rrnS sequence were successful and comparison between this sequence and the annotated copy of rrnS from the complete genome suggest that the preliminary sequence is a pseudogene. The completed genome is 14,908 bp in size and includes all the standard mitochondrial protein coding, rRNA and tRNA genes.
In Coloceras both of the initial long PCR amplicons rrnL to rrnS (CAM10/CAM7, approx. 3350 bp) and rrnS to rrnL (GON1/GON2, approx. 5000 bp) were successful for a mini-circular genome 7649 bp in size that included 22 full length genes, 5 protein coding genes, 2 rRNAs and 15 unique tRNA isotypes plus partial copies of cob (617 vs 1125 bp in full length copy) and nad4 (27 vs 1293 bp) with the intervening genes deleted. Subsequently we amplified a partial cox2 gene (ISC-C2F/ISC-C2R, approx. 280 bp) which we were able to link to cob from the 5’ end (COLO13/ISC-C2R, approx. 1300 bp) and to the control region from the 3’ end (COLO19/COLO22, approx. 6300 bp). The full sized genome is 14,868 bp in size and includes a single copy of each of the canonical 37 mitochondrial genes.
In Anaticola an initial long PCR rrnS to rrnL (GON1/GON2, approx. 2000 bp) and a partial cox1 sequence (L6225/H7005, approx. 380 bp) were generated. These were linked by the amplicon rrnL to cox1 (CAM10/ANAT5, approx. 1500 bp) and extended with the amplicon cox1 to nad4 (ANAT6/N4-J-8924, approx. 2000 bp). A circular molecule was completed by the amplicon nad4 to rrnS (ANAT16/ANAT4, approx. 3500 bp) for a minicircular genome 8,118 bp in size and including 18 genes, 5 protein coding genes, 2 rRNA genes and 11 tRNAs representing 9 different isotypes (3 copies of trnK, 2 sequence identical inverted repeats either side of a non-coding region, and a third which differs by a 8bp indel). In Philopterus both of the initial long PCR amplicons rrnS to rrnL (GON1/GON2, approx. 1600 bp) and rrnL – rrnS (CAM10/CAM7, approx. 2700 bp) were successful for a minicircular genome 3,721 bp in size and including 6 genes, 1 protein coding gene (cox1), 2 rRNAs and 3 unique tRNAs. Subsequently we amplified partial cox2 (ISC-C2F/ISC-C2R, approx. 280 bp) and cob sequences (ISC-CbF/ISC-CbR, approx. 600 bp) but were unable to link either of these sequences either to each other or to the minicircular genome by long PCRs using specific primers designed from each of these sequences. In Quadraceps one of the initial long PCRs rrnS to rrnL (GON1/GON2, approx. 1500 bp) was successful. This sequence was used to design specific primers for the amplicon rrnL to rrnS (QUAD3/QUAD4, approx. 1500 bp), completing a minicircular genome 2,553 bp in size and including 6 genes, 1 protein coding gene (nad6), 2 rRNAs and 3 unique tRNAs. Subsequently we amplified partial cox1 (L6625/H7005, approx 380 bp) and cob (ISC-CbF/ISC-CbR, approx. 600 bp) but were unable to link either of these sequences either each other or to the minicircular genome by long PCRs using specific primers designed from each of these sequences.
In Damalinia, primers designed within the initial, partial sequences of cox1 (COI.out.DS.F/ COI.out.DS.R), rrnL (16S.out.DS.F/ 16S.out.DS.R) and rrnS (12S.out.DS.F/ 12S.out.DS.R), failed to amplify any long PCRs linking these genes. These primers were then used to search for mitochondrial minicircles which include each of these genes by using primers at the 5’ and 3’ ends of each of the partial sequences. Each resulted in a small amplicon of approx. 2000 bp which was sequenced after cloning. Attempts to amplify other mt genes based on cox2 and cob primers conserved across Ischnocera failed. Attempts to design “universal” Damalinia primers based on portions of the non-coding regions conserved across different minicircles (see Fig. 2) also failed to amplify additional genes or minicircles types.