Infection of Mosquitoes with Plasmodium falciparum by Feeding on Humans and on Aotus Monkeys

Patient management.

Consent for whatever treatments the hospital staff determined necessary for the patients was granted by the families of the patients or the courts when patients were admitted to the hospitals. The decision to infect a neurosyphilitic patient with malaria was made as part of standard patient care by the medical staff of the hospital. Patient care and evaluation of clinical endpoints (e.g., fever) were the responsibility of the medical staff. As previously reported,³ during infection, the temperature, pulse, and respiration were routinely checked every 4 hours and hourly during paroxysms (fevers) by hospital personnel. During paroxysms, patients were treated symptomatically. Infections were terminated at the direction of the attending physician. The US Public Health Service personnel provided the parasites for inoculation, monitored the daily parasite counts to determine the course of infection, provided the mosquitoes to be fed on the patients, and performed mosquito dissections and examinations. All patients undergoing malaria therapy were confined in screened wards of the hospital to prevent possible infection of local anopheline mosquitoes.

Treatment.

Patients were infected with one of the strains of P. falcipaum (McLendon,⁷ El Limon,⁸ Santee Cooper,⁹ Colombia, ¹⁰ or Thailand¹¹). Infections were terminated by treatment with various antimalarial drugs appropriate for the particular strain of parasite. Various drugs, such as primaquine, pyrimethamine, quinine, and chlorguanide, are all capable of at least temporarily preventing mosquito infection, without permanently eliminating the infection in the human host. Patients receiving treatments that may have had an effect on mosquito infection were excluded from this analysis.

Blood-stage parasitemia was monitored and quantified by the daily examination of thick and thin blood films by the method of Earle and Perez. ¹² Parasite counts were recorded per microliter of blood but were not included in this report.

Mosquitoes.

Anopheles albimanus mosquitoes were originally from Panama. The An. quadrimaculatus Q-1 strain was originally from the southeastern United States and had been maintained for many years in the laboratory; the An. freeborni F-1 strain was originally from Marysville, California, and has been maintained continuously in the laboratory since 1944. An. freeborni (F-1 strain) were the only mosquitoes selected for analysis for the monkey feeds. For the human studies, mosquitoes were laboratory reared and maintained in the Columbia or Milledgeville insectaries; for the monkey studies, they were raised and maintained in the CDC/DPD insectaries. Mosquitoes were provided with fresh sugar solution daily after their infective blood meal.

Monkeys.

Aotus lemurinus griseimembra were wild caught and imported from Colombia during periods when export was allowed from that country; after that, animals were laboratory bred. Aotus nancymaae and A. vociferans monkeys were imported from Peru; a limited number of these animals were laboratory bred. On arrival at the facility, all animals were quarantined for a 2-month conditioning period, weighed, and tested for tuberculosis. Parasitologic and serologic examination indicated that the animals were free of infection with malaria parasites before inoculation with the Santa Lucia strain, although some of the animals had been previously infected with other species of Plasmodium. All monkeys were splenectomized before exposure. All surgery was performed in an AAALAC (Association for the Assessment and Accreditation of Laboratory Animal Care International)-approved surgical suite appropriate for aseptic surgery. Protocols were reviewed and approved by the Centers for Disease Control and Prevention Institutional Animal Care and Use Committee, in accordance with procedures described in the US Public Health Policy 1986.

Monkeys generally were housed doubly or, in some cases, singly to avoid injuries caused by fighting with cage mates. Space recommendations for laboratory animals were followed as set forth in the NIH Guide for the Care and Use of Laboratory Animals. All animals were fed a diet that has been proven to provide adequate nutrition and calories in captive Aotus monkeys used in malaria-related research. Feed was free of contaminants and freshly prepared. Daily observations of the animals’ behavior, appetite, stool, and condition were recorded. All were treated as medical conditions arose by an attending veterinarian.

During periods when gametocytes were present in the blood, mosquitoes were allowed to feed on tranquilized monkeys. Mosquitoes were contained in pint-sized ice-cream carton cages and allowed to feed directly through netting on the shaved belly of the animal until engorged to repletion. After feeding, mosquitoes were held in an incubator at 25°C until examined ~1 week later for the presence of oocysts on their midguts.

The distribution of the oocysts counts per positive gut [oocysts/(+)gut] was determined at 5-day intervals beginning at the day of maximum asexual parasite count and continuing through Day 30. Included were periods of recrudescence where a secondary maximum parasite count occurred. Calculations were made only for those lots that contained at least one positive mosquito.

Human studies.

An examination was made of mosquito feeding records on patients infected with the four different strains of P. falciparum. The three different species of mosquitoes examined were An. albimanus, An. quadrimaculatus, and An. freeborni. In all cases, the oocyst counts per positive gut [oocysts/(+)gut] for the 1,004 positive lots of mosquitoes that fed on the patients were determined. There were 391 lots of An. albimanus, 488 lots of An. quadrimaculatus, and 125 lots of An. freeborni, which represented the dissection of 24,778 mosquitoes, of which 10,759 (43.4%) were positive.

The results from the feedings on 229 patients are included in the analysis (Table 1). The distribution of oocyst intensity of infection was divided into four groups: 1–10, 10–30, 30–100, and > 100 oocysts/(+)gut.

An examination of this distribution using An. albimanus mosquitoes is shown in Figure 1. Here, the percentage distribution of lots of mosquitoes with values from 1 to 10 (35.4%) and from 10 to 30 (16.6%) = 52.0% compared with the values from 30 to 100 (29.7%) and from > 100 (18.4%) totaling 48%, which indicates that nearly one half of the lots had values of ~30 oocysts/(+)gut for the 391 positive feedings. The average for all the An. albimanus feedings was 55.71 oocysts/(+)gut. Highest level values were obtained during the Day 11–15 period after peak parasitemia (32.0%), closely followed by the 6- to 10-day period after maximum peak positive levels (30.7%). The high values may have been strongly influenced by the fact that most of the feedings were on patients infected with the El Limon strain of P. falciparum originally from Panama, which was also the origin of the An. albimanus. Co-indigenous parasites and vectors often are better suited to each other. ¹³

Examination of feedings with the other two mosquito species, An. quadrimaculatus and An. freeborni (Table 1), indicated higher percentages in the 1–10 oocysts/(+)gut groups and less in the > 30 groups. The An. quadrimaculatus had an average of 25.19 oocysts/(+)gut and the An. freeborni averaged 34.40 oocysts/(+)gut. As indicated, 79.9% with the An. quadrimaculatus feedings were ≤ 30%, and 66.4% in the An. freeborni feedings were ≤ 30. The Day 6–10 feeding period was the best interval with the highest level of oocysts/(+)gut followed by Days 11–15 for all three mosquito species feedings.

Monkey studies.

Mosquitoes were fed on three different species of Aotus (Table 2), of which 94 animals were infective to An. freeborni mosquitoes. There were 64 primary infections: 29 in A. l. griseimembra, 24 in A. vociferans, and 11 in A. nancymaae. Seventeen A. l. griseimembra, 5 A. vociferans, and 3 A. nancymaae had previously been infected with P. vivax. In addition, 5 A. vociferans had been infected with both P. vivax and heterologous strains of P. falciparum.

Aotus lemurinus griseimembra.

Of 794 lots of An. freeborni mosquitoes fed on 29 animals with no previous infection, 348 were infected (43.8%). The average number of oocysts per positive gut was 6.08 (Figure 2). The distribution was 78.7% with 1–10 oocysts/(+)gut, 18.1% with 10–30 oocysts/(+)gut, and 3.2% with 30–100 oocysts/(+)gut. Of 759 lots of mosquitoes fed on 17 animals previously infected with P. vivax before being infected with P. falciparum, 451 were infected (59.4%). The average number of oocysts per positive gut was 5.68 (Figure 3). The distribution was 81.4% with 1–10, 15.1% with 10–30, and 3.5% with 30–100 oocysts/(+)gut. It was readily apparent that, compared with the human feedings, the level or intensity of infection was much lower. However, the time pattern of infection was similar to humans with highest level of infections occurring between Days 11 and 15 (24.4%). Previous infection with P. vivax did not seem to diminish the intensity of infection nor alter the days when the highest level infection occurred.

Aotus vociferans.

Of 427 lots of An. freeborni mosquitoes fed on 24 animals with no previous infection, 241 were infected (56.4%). The average number of oocysts per positive gut was 2.78 (Figure 4). The distribution of oocysts per positive gut was 92.5% with 1–10, 6.6% with 10–30, and 0.8% with > 30. These were lower than obtained with the A. l. griseimembra monkeys.

Aotus nancymaae.

Of the 301 lots of An. freeborni mosquitoes fed on the 14 A. nancymaae, 151 were infected (50.2%). The average number of oocysts per positive gut was 10.40. The distribution of the oocysts per positive gut values for An. freeborni fed on A. nancymaae is shown in Figure 5. The distribution of oocysts per positive gut was 61.3% with 1–10, 28.7% with 10–30, 10.0% with 30–100, and one had a value > 100 (0.7%). It was apparent that feedings on A. nancymaae resulted in the highest oocysts counts per positive gut.

The distribution and intensity of infections obtained by an examination of the 1,191 positive infections (Table 2) indicated that the feeding of An. freeborni on the Santa Lucia strain of P. falciparum in all Aotus species tested resulted in 80.3% of the lots being infected at a level ≤ 10 oocysts/(+)gut; 16% were infected at a level of 10–30 oocysts/(+)gut. The highest oocyst counts were obtained during Days 11–15, followed by the feeding periods of Days 6–10 and 16–20.