A RETROSPECTIVE EXAMINATION OF MOSQUITO INFECTION ON HUMANS INFECTED WITH PLASMODIUM FALCIPARUM

INTRODUCTION

Jeffery and Eyles¹ reported on the infectivity to mosquitoes of Plasmodium falciparum as related to gametocyte density and duration of infection. Mosquitoes frequently became infected on gametocyte densities < 10/μL of blood and were infected as late as day 410 of patent parasitemia. It was concluded that in certain endemic areas, long, enduring gametocytemia may be responsible for a large part of the transmission of this parasite. In their studies, mosquitoes were fed on 59 infections with the El Limon (Panama) strain and 29 infections with the Santee Cooper (South Carolina) strain of the parasite. The Q-1 strain of Anopheles quadrimaculatus and two strains of An. albimanus were used. They concluded that the number of lots infected was in general directly proportional to the gametocyte density. They pointed out that there was a period early in the infection when gametocytes were present, but infection did not occur. They also noted that the highest rate of infection of lots occurred during the first 30 days.

The patterns of mosquito infection vary between the different species of Plasmodium infective to humans. With P. vivax and P. ovale, mosquito infection often coincides with the asexual parasite count. However, with P. falciparum, mosquito infection is often markedly delayed, corresponding to the appearance of gametocytes, which usually occurs a week to 10 days after the detection of asexual parasites in the peripheral blood.

As part of our continuing efforts to search archival records on induced infections in patients for the treatment of neurosyphilis between 1940 and 1963, an examination was made of levels of mosquito infection as related to the appearance and number of gametocytes present in the peripheral blood. Mosquitoes had been fed to provide sporozoites for subsequent passage to other patients.

Reported here are the results of a retrospective examination of archival data from 173 induced infections with P. falciparum for the treatment of paresis and other mental disorders associated with tertiary syphilis. The goal was to further document the relationship between gametocyte count and 1) the percentage of mosquito lots infected, 2) the percentage of mosquitoes within lots being infected, and 3) the intensity of mosquito infection (oocysts per positive gut).

MATERIALS AND METHODS

RESULTS

The data available are represented by three patients: G-48, G-120, and G-139. Patient G-48 was infected via the intravenous injection of erythrocytes infected with the Santee Cooper strain of P. falciparum (Figure 1). No fever chart is available. Anopheles quadrimaculatus mosquitoes were fed and examined for the presence of oocysts. Gametocytes first were observed on day 10 of patent parasitemia and mosquito infection was first obtained on day 14. Peak mosquito infection occurred on days 23, 31, 36, and 50. Mosquitoes were infected continuously from days 14 through 59.

Patient G-120 was also injected with infected erythrocytes with the Santee Cooper strain (Figure 2). Fever was recorded on seven different occasions between days 7 and 39 of patent asexual parasitemia. Gametocytes were first observed on day 13, at which time feeding of An. quadrimaculatus commenced. Between days 13 and 80 of patent parasitemia, mosquitoes were fed and subsequently examined for the presence of oocysts on 35 different occasions. Every lot of mosquitoes was infected.

Patient G-139 was infected with the Santee Cooper strain via the bites of infected mosquitoes (Figure 3). Fever was recorded on the first day of patent parasitemia, 9 days after infection. Fever continued through day 15. Gametocytes were first observed on day 11, and mosquito infection commenced on day 18 of patent parasitemia. Between days 18 and 76, 25 of 26 lots of An. quadrimaculatus were infected.

A total of 913 lots were fed on 173 patients (Table 1). Lots fed during the first 80 days of gametocytemia averaged greater than 50% infected; the mean percentage of mosquitoes infected in these lots was slightly less than 50% (Figure 4). Periods of detectable gametocytemia were often interspersed with undetectable periods; mosquito infection continued to occur in only a few patients beyond 200 days.

An examination of the data indicated that the primary period of mosquito infection occurred during the first 20 days of gametocytemia (Figure 5). The peak geometric mean gametocyte count in patients on whom mosquitoes were fed was 2,265/μL on the seventh day of gametocytemia. For those lots that were infected during the first 30 days of gametocytemia, 109 occurred when the count was > 1,000/μL and 88 occurred when the count was between 100 and 1,000/μL. During the first 20 days of gametocytemia, 209 (67.20%) of 311 lots of mosquitoes were infected, and of these, 163 had greater than 50% of the mosquitoes in the lots infected with at least one oocyst.

Many patients were allowed to have extended periods of parasitemia that resulted in secondary waves of asexual parasitemia, followed by secondary waves of gametocytemia. During such periods, (Figure 6), 293 (78.76%) of 372 lots of mosquitoes were infected. The highest percentages of mosquitoes were infected from four days before to four days following the peak gametocyte density.

In addition to the percentage infection, an examination was made of the intensity of infection as related to the gametocyte count. The mean number of oocysts per 100 gametocytes in 408 lots of mosquitoes fed on patients was determined (Figure 7). The ratio of oocysts to 100 gametocytes/μL was greater after day 50. However, at this time, feedings were more likely made on infections in patients previously proven to be infective.

DISCUSSION

Infections with P. falciparum readily infected An. quadrimaculatus and An. albimanus mosquitoes over extended periods of time. Thus, a continuous infection with this parasite would present abundant opportunities for infection and subsequent transmission. The implication for the field is that failure to cure an infection with P. falciparum may result in high levels of mosquito infection during recrudescent periods when gametocytes may be abundant in the absence of clinical malaria. It was apparent that a higher gametocyte count was more apt to result in infection, at least during the first 30 days of patent gametocytemia. However, lower density gametocytemia following the clinical attack, often resulted in mosquito infection.

Fever in many patients was concentrated in the first two weeks of patent parasitemia. This was followed by the first wave of gametocytemia, during which time mosquito infection occurred. Fever was seldom evident during subsequent increases in asexual parasitemia and gametocytemia. However, during this time, mosquito infection frequently occurred. There appeared to be no direct relationship between fever and mosquito infection, but a direct relationship to the presence of gametocytemia.

The gametocyte to oocyst ratio was markedly low. In previous studies,⁴ the average blood volume in a meal was determined to be 3.30 μL for An. quadrimaculatus and 2.44–2.46 μL for An. albimanus. Thus, the mosquitoes would have ingested 2–3 times the number of gametocytes/μL. Many mosquito species may be inefficient in supporting maximum development of the potential number of oocysts (as indicated by the number of gametocytes available). Many factors may contribute to this. One might be a distorted ratio of microgametocytes to macrogametocytes. Another might be due to a difficulty on the part of the ookinetes in penetrating the peritrophic membrane surrounding the blood meal in these mosquitoes. It is also possible that other species of Anopheles would have had more or fewer oocysts than were produced in An. quadrimaculatus and An. albimanus. Both of these mosquitoes were proven hosts and vectors of the P. falciparum strains studied. Disparities in infection rates for different vectors fed on different isolates of Plasmodium are well recognized. Unfortunately, the Santee Cooper, Panama, and McLendon strains of P. falciparum are no longer available for introduction into experimental hosts such as Aotus monkeys. With monkey-adapted strains, comparative infection rates for these and other species of Anopheles can vary markedly, depending on the strain of parasite.

From these data it is apparent that infections with P. falciparum in humans are readily infective to these anopheline mosquitoes, and that higher numbers of mosquitoes are usually infected when gametocyte counts are highest. Mosquito infection could be obtained continuously during the early stages of infection, and, at times, could continue for extended periods in some patients.

These findings are similar to mosquito infection rates obtained in studies with Aotus monkeys experimentally infected with P. falciparum.⁵ In monkeys that have been splenectomized, patterns of infection indicate that successive waves of asexual parasitemia are often followed by secondary periods of mosquito infection (Figure 8), as was observed in humans. In contrast to humans, splenectomized Aotus monkeys demonstrate immature gametocytes in the peripheral circulation. In monkey studies, gametocyte to oocyst ratios have been very low.