When I was an undergraduate biology student, one of my favorite topics was the complement system in immunology. The complement cascade is an array of sequentially interacting proteins that serve a vital role in innate immune responses. The complement cascade can be activated via interactions with antibody-antigen complexes. Proteins involved in the complement cascade react with one another and with components of the target cell, marking pathogen cells for recognition by phagocytes or inducing cell membrane damage, leakage of contents, and cell lysis. The accompanying animation shows the formation of the membrane attack complex, which serves to punch a hole in the cell membrane, resulting in cell lysis and death.

The complement cascade needs to be very finely regulated to prevent damage to self-cells by antibody-directed complement-mediated lysis. Further, the complement cascade needs to be controlled because degradation products of the complement proteins can diffuse (and thereby cause damage) to adjacent cells. The complement cascade is thus very tightly regulated by several circulating and membrane-bound proteins.
There are three major pathways of the complement system. These are the classical pathway, the alternative pathway and the lectin pathway. To give a sense of the complexity and engineering brilliance of the complement cascade, let me briefly describe the classical pathway.

The first stage is the initiation phase, and the classical pathway is triggered by antibody molecules bound to antigens. An enzyme called C1, found in blood serum, has an affinity for immunoglobulins. C1 is a molecular complex comprised of 6 molecules of C1q, 2 molecules of C1r, and 2 molecules of C1s (C1qr2s2). The constant regions of mu chains (IgM) possess a C1q binding site. Some gamma chains (IgG) also possess this binding site but IgG is much less efficient than IgM, and many molecules are needed to initiate the pathway (whereas only one molecule of IgM is required).

Since C1 can readily undergo autoactivation, it is ordinarily regulated by a C1-inhibitor protein (C1-In or C1 esterase). This inhibiting activity, however, is overcome upon binding of immunoglobulin molecules to C1q. Upon binding of activators to C1q, the C1r and C1s components of the C1 molecule are activated (C1r* and C1s*), and they are rendered catalytically active.
Two serum proteins, C4 and C2, are cleaved by C1s*. C4 is cleaved to form C4a and C4b. C4a has no further use and diffuses away, while C4b covalently binds to transmembrane glycoproteins. C2 is cleaved into C2a and C2b. C2a has no further use and diffuses away. C2b binds to C4b. By convention, the larger subcomponent is always designated “b” and the smaller subcomponent is designated “a.”

The complex that is formed by this association between C2b and C4b is responsible for catalyzing the cleavage of C3, and thus it is named the C3 convertase (C4b2a). C3 is cleaved into C3a and C3b. C3a diffuses into the plasma. When C3b joins the C3 convertase, it forms the C5 convertase (C4b2a3b). The C5 convertase subsequently cleaves protein C5 to form C5a and C5b. C5a diffuses into the plasma, but C5b is responsible for initiating the formation of the membrane attack complex (MAC). The membrane attack complex is assembled by C5, C6, C7, C8, and C9. As many as 18 C9 molecules form a tube that is inserted into the membrane, creating a transmembrane channel. Water osmotically enters the cell, causing it to burst.

There is much more detail that could be given, of course. And I haven’t even touched on how this cascade is regulated (which involves many other proteins). It is extremely difficult to envision how an ordered (and tightly regulated) cascade or pathway, such as complement, could have arisen in step-wise Darwinian manner. But these are precisely the types of systems that are created by intelligent agents. The more we learn about biology at a micro scale, the more clearly it manifests design.