Five of 26 flathead mullet (Mugil cephalus) captured in the Seekoei River estuary in South Africa showed haemogregarine infections. Both trophozoites and gamonts were observed, and image analysis technology was used in parasite morphometric description. Gamonts were compared with those of Desseria mugili from Brazil and found to be different. The parasite mostly closely matched a haemogregarine previously illustrated, but not described from South Africa, and apparent lack of merogony suggested that it was a species of Desseria.

 

The most commonly reported intraerythrocytic apicomplexan from mullets is Desseria (Haemogregarina) mugili (Carini, 1932) Siddall 1995. Also recorded from mullet blood cells are other types of haemogregarines (Fantham, 1919; Saunders, 1955; Al-Salim, 1989; Paperna, 1996) and a dactylosomatid (Paperna, 1981). This short communication records an intraerythrocytic haemogregarine from Mugil cephalus L., captured in South Africa, and the use of image analysis in its morphometric description. The parasite is compared with D. mugili from Brazilian mullet (see Eiras, Ranzani-Paiva & Davies, 1995) and found to be different.

South African material was collected in January 1998, 10 km west of Jeffreys Bay in the Seekoei River estuary. Twenty six immature flathead mullet (M. cephalus), were caught with cast nets in 1.4 m of brackish water, measured (total length), and then killed with benzocaine (ethyl-4-aminobenzoate).  Heart blood was smeared on clean glass slides, fixed in absolute methanol and stained with Giemsa’s stain.  Smears were screened with a 100 X oil-immersion objective using a Zeiss Axiophot 20 photomicroscope. Parasite images were captured by digital camera (Nikon DN100), stored on computer disks and then measured with an Eclipse Net (Nikon) image analysis package, calibrated to a stage micrometer. Blood smears taken from Brazilian mullet (see Eiras et al., 1995) were re-examined with the same equipment for comparison with South African samples.

In blood smears taken from Seekoei River catches, three distinct intraerythrocytic stages of a haemogregarine were found in 5/26 (19.2%) flatheads, measuring 154.0 ±40.3 (range 96−251) mm long. In 1/5 flatheads with haemogregarines, a light infection with a trypanosome was also noted and this will be described elsewhere. Intensity of infection with haemogregarines was between 1/500 (0.2%) and 1/10000 (0.01%) of erythrocytes infected, all forms occurring singly within erythrocytes and none undergoing division. Hosts cells appeared relatively unaffected by the haemogregarines.

The smallest intraerythrocytic stages, possibly trophozoites, were pale-staining, measuring 5.7 ±0.5 (5.0−6.7) mm long by 2.0 ±0.4 (1.6−2.6) mm wide (N = 15) (Figures 1A,B; 2A,B). They were elongate, occasionally with one end slightly broader than the other.  Their nuclei (1.6−3.1 mm by 0.8−2.0 mm), were centrally placed (Figure 1A) or tended to occur towards one end (Figures 1B; 2A,B), and appeared to comprise a number of reddish staining granules.  Cytoplasm stained pale blue, and one or two vacuoles and granules were present.

Immature gamonts (Figures 1C; 2C) measured 7.4 ±0.4 (6.8−7.9) mm long by 2.1 ±0.1 (1.9−2.3) mm wide (N = 15). They were also elongate, with the presumed anterior end only slightly broader than the posterior, and the nucleus often situated centrally. Nuclei measured 2.7−3.8 mm by 1.8–2.2 mm, with chromatin staining reddish-purple. Cytoplasm stained deeper blue than that of the earlier stages and no vacuoles were detected. 

Monomorphic, mature gamonts (Figures 1D–F; 2D–F) were 8.8 ±0.8 (7.5–10.9) mm long by 2.5 ±0.4 (1.7–3.2) mm wide (N = 25). These mature stages were broad and elongate, with their extremities slightly curved towards the nucleus of the erythrocyte. In all specimens, the presumed anterior end was broader than the somewhat pointed posterior.  Nuclei were well defined, with condensed chromatin that stained deep purple, and were situated centrally or in the posterior two thirds of the gamont. Nuclei measured 3.3 ±0.4 (2.8–4.3) mm long by 1.7 ±0.2 (1.1–2.0) mm wide (N = 25). Cytoplasm stained deep blue and contained characteristic dark-staining granules, especially near the anterior end of the gamont (arrows Figures 2 D,E). A few vacuoles occurred between the nucleus and the anterior end.

Image analysis technology proved a rapid and accurate means of measuring parasites found in mullet blood samples from South Africa. Furthermore, when parasites from Brazil were re-measured using this package, the length and width ranges of D. mugili from this source were found to match those reported by Eiras et al. (1995) using standard techniques.

Further studies on flathead mullet from the Seekoei River estuary in a variety of seasons are needed to establish the firm identity of this intraerythrocytic apicomplexan, although some comments on its likely identity are possible. Merogony was not detected which suggests that it is a species of Desseria (see Siddall, 1995). However, although the dimensions of its mature gamonts only marginally exceed those reported for D. mugili (see Eiras et al., 1995), there are marked differences in morphology.  Gamonts with condensed nuclear chromatin and prominent anterior granules exist in mullet from South Africa, but gamonts of D. mugili have nuclei with chromatin blocks and lack anterior granules (Figures 1G,H). Trophozoites, but no cresent forms of the haemogregarine were detected in Seekoei River mullets during the current study. In contrast, Eiras et al.’s (1995) blood films from Brazil lack trophozoites, but have crescent stages up to 18.3 mm long (Figure 1I). Effects on host cells between South African and Brazilian material also differ, the latter erythrocytes often exhibiting rectangular nuclei when infected (see Eiras et al., 1995). Given these differences, it is impossible to identify the parasite from the Seekoei River as D. mugili.

The apparent absence of merogonic stages also separates the South African haemogregarine from Haemogregarina bigemina Laveran & Mesnil 1901, which Saunders (1955) recorded from mullet in the Bahamas, and from Haemogregarina meridianus Al-Salim, 1989 from mullet in Iraq (Al-Salim, 1989). In addition, the parasite does not match Fantham's (1919) brief report of a haemogregarine from a mullet captured in Kalk Bay, South Africa. Fantham described a globular haemogregarine and a rounded meront with six nuclei, three of which were undergoing a second division. This is similar to Paperna’s (1981) reported development of Babesiosoma (Dactylosoma) hannesi (Paperna, 1981) Barta 1991 from a variety of mullets from the Swartkops estuary, east of Port Elizabeth. However, the gamonts of Paperna's dactylosomatid, are approximately half the length of gamonts reported here and their staining characteristics are very different.

One possible match remains for the haemogregarine from the Seekoei River.  Paperna (1996) illustrated, but did not name, a single gamont (?) of a haemogregarine from the blood of L. richardsonii.  The gamonts of our haemogregarine most closely resemble the parasite in Paperna’s micrograph. Although we are reluctant to name the parasite from South Africa as a new species, it can be provisionally placed in the genus Desseria.