As we mentioned last meeting, insects defend themselves from challenges by microorganisms with a combination of humoral and cellular responses. The distinction between these two response categories is rather artificial for a couple of reasons. First, hemocytes can release some of the protein factors that follow from challenges. Hemocytes produce what we regard as humoral responses. Second, although infections elicit a complex of multifaceted responses, the responses are coordinated and integrated into a single theme: host defense. Nevertheless, it is very convenient to treat hemocytic and humoral immunity as separate topics. One point of separation is that hemocytic responses are rapid. They occur almost immediately an insect is invaded by microorganisms. Humoral responses are much slower, they are usually not seen until some hours after infection. These longer-term responses may be more important in sustaining the hemocytic host defense reactions initiated earlier in the infection cycle.

Circulating cells are regularly found in the hemolymph of insects. These cells are known as hemocytes. Hemocytes are involved in conferring a cellular, or hemocytic, immunity to insects. These cells are formed in haemopoietic organs that are variously well developed in different insect groups. In the cricket Gryllus haemopoietic organs are well developed, distinct organs found in the second and third abdominal segments along either side of the heart. In other insects haemopoietic organs are amorphous clumps of cells irregularly distributed along the dorsal vessel.

You might enjoy a quote on the classification of insect hemocytes: "Classification of insect hemocytes is an issue that stimulates an instantaneous loss of good manners and an induction of bad temper at any conference on insect physiology or immunity" . {Lackie, A.M., 1988, In: Advances in Insect Phyisology, Ed. by P.D. Evans and W.B. Wigglesworth, Academic Press, London} Given detectable levels of disagreement on the subject, a list of some hemocytes and their function follows:

Prohemocytes (PRs) are thought to be the stem cells from which all hemocytes arise.

Plasmatocytes (PLs) are larger cells involved in major immunological actions. PLs are capable of phagocytosis, a kind of entocytosis of smaller foreigh cells. Phagocytosis usually follows low doses of invading yeast and bacteria. PLs are also involved in encapsulation of clumps of invading microbes. There are typically more PLs than other types of hemocytes.

Granulocytes (GRs) are characterized by the presences of many intracellular granules. Electron microscope studies have revealed differences among the granules in these cells. There are structureless and structured granules. The biochemistry of the different types of granules is not yet understood. GRs are major players in encapsulation reactions.

Other hemocytes include spherulocytes, oenocytoids and adipohemocytes. These cells are regarded as "nonimmunocompetent" by some authors, although there is considerable ground to disagree with this idea. Gupta prefers to use the term "immunocytes" for GRs and PLs and "hemocytes" for all the other types. He uses these terms to stress the idea that GRs and PLs make up most of the immune responses. Spherulocytes may be involved in silk formation, melanization and secretion of some hemolymph proteins. They may also be important in hemolymph clotting responses to wounds. Oenocytoids may contribute to the final stages of cuticle formation. Other authors suggested that these cells may recognize "non-self".

Again, hemocytic defense responses occur immediately insects are invaded by microorganisms. GRs and PLs are responsible for a number of cellular immune reactions. They are able to phagocytize small biologic particles such as bacteria and yeast. Phagocytization refers to the process of cellular internalization of foreign invaders. Phagocytosis is not the same as killing the microbial cells. Microbes are secondarily killed after phagoccytosis.

GRs and oenocytoids are attracted to sites of wounds, where they play major roles in coagulating hemolymph to stop bleedings. We know virtually nothing about the signalling process that is responsible for moving cells to wound sites.

Another defense mechanism is nodule formation. Nodules are aggregates of hemocytes that entrap invading microbes. This is a particularly effective means of clearing the hemolymph, and it is considered to be of greater importance than phagocytosis in clearing large doses of bacteria. There appears to be a sequence of trapping by GRs, followed by aggregation by PLs. The precise sequence of events will remain a point of controversy. In addition to bacteria, viral polyhedrons, fungal spores and protozoans are taken up in nodules.

When organisms or clumps of organisms are too big to be phagocytized or to form nodules, they may be encapsulated. Encapsulation is the sequesterization of these organisms or clumps with multilayered aggregates of hemocytes. This also involves release of coagulum, in a response similar to wound-repair. The coagulum probably aides in forming layers of cells. Encapsulation was first observed by the Russian zoologist Elie Metchnikoff in 1892. He placed rose thorns into a starfish larva. The next day he saw that the thorns were completely surrounded by tiny cells. Metchnikoff called these cells phagocytes from the Greek words for eating and cells. Metchnikoff eventually moved from the University of Odessa to the Pasteur Institute, and later shared the Nobel Prize for his studies of cellular defenses in mammals and invertebrates.

We are not sure of the killing mechanisms that hemocytes elaborate. Some of the suggested mechanisms include peroxidation, melanization, and lysozymes. By comparison to our scant knowledge of invertebrate immunity, very much is understood about the killing mechanisms of mammalian host-defense cells. There are two sorts of mechanisms. One sort does not depend on oxygen. These include protease, esterase, acid-phosphatase and lysoszyme mechanisms. Another sort does depend on oxygen, including formation of singlet oxygen and superoxide anions, which are very reactive and very toxic species. Work on killing mechanisms of insect hemocytes will produce important and very novel insights into how insects protect themselves.

We recently addressed the question of how the early, hemocytic responses to infection are mediated. What we wanted to know was what biochemical signals occured between recognition of a bacterial infection and the hemocytic responses that follow. Since many aspects of host defense in mammals are mediated prostaglandins and related eicosanoids, we formed the hypothesis that these molecules are also crucial mediators of immunity in insects. The central experimental design was to inhibit various of the eicosanoid biosynthetic pathways, then observe the effects of inhibition on the insects' ability to remove bacterial infections from their hemolymph.

Since hemocytic responses occur in the first few minutes post-infection (PI), we could dissect the humoral responses from hemocytic responses by doing the experiments within the first hour PI.

In the first experiments, cells of the insect pathogenic bacterium Serratia marcescens were injected into 5th instar larvae of the tobacco hornworm Manduca sexta. Test animals were first injected with the phospholipase A2 inhibitor dexamethasone, and control animals were first treated with pure ethanol. Hemolymph samples were withdrawn at 15, 30 45 and 60 minutes PI. The samples were diluted slightly, then streaked onto agar plates. The plates were incubated for about 40 hours, then the number of red colonies on each plate were counted.

Until these data are scanned into the Web site, you'll have to look at the original publication, or use the handouts. The line connected by square boxes represents hemolymph samples from control insects: there were no red colonies on these plates, indicating that the insects had cleared all or most of the bacteria from their hemolymph. The line connected by X's represents hemolymph samples from test insects. As early as 15 minutes PI, there was a statistically significant increase in recovered bacteria. The number of recovered colonies increased over the first hour PI.

We tested the effects of several doses of dexamethasone by injecting 4 doses in different groups of insects. The results indicate that more bacterial colonies were recovered with increasing doses of dexamethasone.

The bacteria were used in these experiments are insect pathogens. We reasoned that the dexamethasone decreased the ability of the larvae to clear bacteria from their hemolymph. Since these bacteria are pathogens, the decreased clearance from hemolymph should translate into increased mortality from these infections. We tested this idea by injecting the same 4 doses that were just described into 4 groups of larvae. After 14 hours we counted the number of surviving larvae. As expected, increasing dosages of dexamethasone were associated with decreasing larval survival at 14 hours.

Dexamethasone inhibits the activity of phospholipase A2. This in turn inhibits eicosanoid biosynthesis because substrate is not available to the eicosanoid-biosynthetic pathways. We conducted experiments to test the possibility that arachidonic acid would rescue the larvae from the effects of dexamethasone. These kinds of experiments are called "end product reversal experiments", or something like that. The last bar in the figure above this paragraph shows that when larvae were treated with dexamethasone, and with arachidonic acid, larval survivorship was a great as survival in the control group.

We can see from the first figure that phospholipase A2 activity is necessary for all eicosanoid biosynthesis. We wondered if one eicosanoid biosynthetic pathway might be more important than another in mediating insect immune responses. To investigate this point we treated groups of larvae with specific inhibitors of phospholipase A2 (dexamethasone), cyclooxygenase (indomethacin), lipoxygenase (esculetin), and indomethacin and esculetin together. We also observed the effects of maleic acid, an antioxidant compound that does not interfere with the major eicosanoid biosynthesis pathways. After treating the larvae with inhibitors, we infected them with bacteria. After 2.5 hours, we withdrew hemolymph samples, and plated them on agar plates. We then counted the number of bacterial colonies on each plate. This figure shows the proportion of larvae from which ANY colonies were recovered. This is a rigorous experiment because if any bacterial colonies were recovered, even a single colony, the individual was counted as positive. These data show that at least 1 colony was recovered from about 20% of the control larvae. Colonies were recovered from about 80% of the phospholipase A2-inhibited larvae. Bacterial colonies were also recovered from the cyclooxygenase-inhibited and the lipoxygenase-inhibited larvae, but from about 60%, rather than the 80% of phospholipase A2-inhibited larvae. These findings indicated that both pathways were involved in the immune response to infections. This idea was supported by the results from the group that was treated with both inhibitors at the same time: they were not different from the dexamethasone effects.

We concluded that products of cycloxygenase and lipoxygenase pathways were involved in mediating cellular responses to bacterial infections in this insect.

The last comment, of course, opens questions of identifying specific cellular actions that are mediated by eicosanoids. As mentioned earlier, nodule formation is responsible for clearing the bulk of bacterial infections from circulation. We developed the hypothesis that eicosanoids mediate the nodulation reaction to bacterial infections. Using a line of research similar to the one just described, we found that treating tobacco hornworms with eicosanoid biosynthesis inhibitors just prior to infecting them with bacteria resulted in severe impairment of nodulation. The influence of the inhibitors is expressed in a dose-dependent manner and can be reversed by eicosanoid-precursor polyunsaturated fatty acids. Again, we concluded that products of cycloxygenase and lipoxygenase pathways were involved in mediating nodulation responses to bacterial infections in this insect.

Finally, we began testing the broader hypothesis that eicosanoids mediate nodulation in other insect species. So far, we have tested these species:

This work serves to support the broader hypothesis that eicosanoids mediate nodulation responses in all insects that defend themselves from bacterial infections by forming nodules. Since publishing our original findings, other laboratory groups have taken up this line of work. We now know that eicosanoids mediate phagocytosis and nodulation in greater waxmoths.