I am deeply honoured to have been elected doctoris honoris causa at the Univer
sity o f Gdańsk and I w ill now te ll a litde about myself, the science I have been involved in , and about the marvellous cooperation I and m y co-workers have had w ith great Polish scientists m ainly associated w ith the U niversity o f Gdańsk.
I became very interested in chem istry when I was about ten years old human chem istry is the m ost complex one and that it produces both health, disease, and consciousness. E qually fascinating was the in sig h t that all o f this was guided by the properties o f the elements, w hich in th e ir tu rn were inherent in the B ig Bang event. Since the B ig Bang was soundless, it m ight be more adequate to name it the B ig Brainwave o r M in d Wave.
So, w hen in the early 1970s I arrived at the intersection o f chem istry and human heakh and disease, called C linical Chemistry, I was eager to start exploring the w o rld o f human chemistry. M y firs t encounter w ith a molecule o f unknow n structure and function resulted in m yself being im printed by it, ju st like some birds, according to Lorenz, get im printed by the firs t creature they see when they jum p o u t o f th e ir eggs.
T his was the p ro te in cystatin C, w hich in these days had ju s t a triv ia l name, since v irtu a lly nothing was know n about its structure o r function or possible use in medicine. So, we isolated it and sequenced its single polypep
tide chain, w hich in these days was n o t easy. We then hoped d ia t we w ould get help fro m the developing com puter science as this allowed comparison o f all am ino acid sequences o f all proteins w ith know n and unknown func
tions. For i f a sequence hom ology between cystatin C and a protein w ith know n fu n ctio n w ould be found, it w ould strongly indicate that cystatin C had a sim ila r fu n ctio n . B u t the cystatin C sequence d id n o t resemble any other know n sequence. This was bad, because it meant that we had no clue about its function, b u t also good, because it meant that cystatin C was the firs t member o f a new and exciting superfam ily o f proteins.
57
The possibility o f com paring a new am ino acid sequence w ith all previ
ously know n am ino acid sequences to gain in fo rm a tio n reflects the im p o r
tant fact that m odern chem istry represents a universal language understood by chemists from all countries in the w orld. The many languages o f Earth use a m ultitude o f different letters, but the letters o f chem istry are just the desig
nations o f the elements. These designations were suggested by the Swedish physician and chemist Jons Jacob Berzelius in the beginning o f the 19* cen
tu ry and meant that all elements should be designated by the first, o r the tw o first, letters o f their La tin o r Greek names. So, since about 200 years and fo r all eternity all chemists in all countries w ill know exactly what C, N , O, H , and H ,0 mean, but only a fraction o f all people w ill know w hat, fo r instance, is the meaning o f “ navcaq” .
The universal language o f chem istry allowed the discovery o f the func
tio n o f cystatin C by connecting it to tenderizing meat. In C alifornia meat is often tenderized by inje ctio n o f papain in to the animals at slaughter, and the cattle breeders were annoyed that so m uch expensive papain was required.
They therefore supplied grant money to nearby universities w ith the hope th a t the researchers w o u ld be able to explain the vast amounts o f papain required fo r tenderizing meat. Some researchers coupled papain to an insol
uble m atrix in a colum n and poured diffe re n t biological fluids through the colum n and investigated i f anything w o u ld stick to the papain. W hen egg w hite was poured through the colum n, a protein bound to the papain and could be isolated and sequenced. Its sequence was highly homologues to that previously determ ined fo r cystatin C and we could im m ediately establish that cystatin C was an in h ib ito r o f papain. I t is now know n that cystatin C is the m ost im p o rta n t in h ib ito r o f many v ita l human cysteine proteases, fo r example cathepsins B, K , L , and H .
One interesting aspect o f proteins is that th e ir function is often executed by a very small part o f them called the active site or, fo r protease in h ib ito rs, the in h ib ito ry site. This means that i f you can fin d the site, it m ig h t be pos
sible to produce low -m olecular-w eight substances w ith the same, o r sim ilar, biological effects as the big protein. These low -m olecular-w eight substances can often be used as medicines. However, identification o f the active o r in h ib ito ry site(s) o f a protein is m ost often very d iffic u lt and requires knowledge o f the three-dim ensional structure o f the protein. B ut fo r cystatin C we made the serendipitous observation that some preparations o f cystatin C contained a sm all fra ctio n o f cystatin C in w hich the eleven W -term inal am ino acid residues were m issing. This variant had lost m ost o f the in h ib ito ry capacity o f cystatin C and it was therefore very like ly that at least a part o f the in h ib ito ry site o f cystatin C was constituted by a part, o r all, o f the N -te rm in a l undecapeptide. This observation was done in 1986 and we wanted to test i f
58
peptidyl derivatives structurally based upon this undecapeptide could m im ic the properties o f cystatin C fo r example as a cysteine protease in h ib ito r.
We trie d to fin d scientists o r chemists in Sweden w ho w ould be able to syn
thesize such peptidyl derivatives, but, despite considerable efforts, we were n o t able to fin d any. Finally, we contacted a chem ist, Jerzy Trojnar, in the Ferring company in M alm o. H e originated from Poland and knew Zbigniew Grzonka and his group at the U niversity o f Gdańsk and he to ld me that i f any
one w ould be able to synthesize the required peptidyl derivatives, it w ould be Z bigniew and his research group. Jerzy contacted Z bigniew and established the lin k between him and me. For young people, constantly hugging their m obile phones, it m ig h t be d iffic u lt to understand how problem atic it was to com m unicate between Poland and Sweden in the 1980s. For every phone ra il Z bigniew and I were going to have, we had to order it at least fo u r hours in advance so th a t a suitable censor could be selected to check that Z bigniew and I were n o t conspiring against the com m unistic governm ent o f Poland.
These censors were hard ly chemists and m ust have been bored to death by our chem istry discussions. B ut th e ir basic assumption was valid; neither Z bigniew no r I appreciated the com m unistic governm ent very much. N a tu rally, Z bigniew and his co-workers, among them Franciszek Kasprzykowski and Leszek Pankiewicz, were able to synthesize interesting peptidyl deriva
tives, m im icking the protease in h ib itin g effects o f cystatin C. As a m atter o f fact, some o f them were, and s till are, the best cysteine protease in h ib ito rs available. As often occurs in chem istry and science, some o f the p e p tid yl derivatives showed surprising addition al properties and turned o u t to be efficient a n tivira l and antibacterial substances. They could even lu ll bacterial strains resistant to all know n antibiotics. The fu rth e r developments in this fie ld have m ainly taken place in Gdańsk w ith die aim o f developing new types o f antibacterial and antiviral substances. The firs t product fo r medical use has recently been introduced in Gdańsk.
B ut, although the N -term inal part o f cystatin C clearly was involved in the in h ib ito ry site o f it, a fu ll characterization o f the site required knowledge o f the three-dim ensional structure o f cystatin C. We had determ ined the nucleic acid sequences o f the m R N A and gene fo r cystatin C and were able to con
struct a system fo r efficient production o f recom binant cystatin C, identical w ith native cystatin C isolated from human fluids. O u r research group d id not have the capacity to use crystallography in our protein studies, so we con
tacted crystallography groups all over the w o rld to try to interest them in the three-dim ensional structure o f cystatin C. They were all very interested and we sent several grams o f recom binant cystatin C to them fo r these studies.
However, although tw o groups were led by N obel Prize laureates, nobody could present data on the three-dimensional structure o f cystatin C. However,
59
Z b ig n ie w in vite d co-workers, experienced in crystallography, fro m Adam M ickiew icz U niversity in Poznan, am ong them M ariusz Jaskolski, R obert Janowski, and M aciej Kozak, whose innovative th in k in g made it possible to determ ine the three-dim ensional structure o f crystalline cystatin C to be a domain-swapped dimer. This was a m ajor step in defining the in h ib ito ry site o f cystatin C, but also supported a new hypothesis on how native proteins m ig h t aggregate and form the detrim ental fib rilla ry deposits o f am yloid dis
orders, like Alzheim er’s and Parkinson’s diseases. Since cystatin C is norm ally present as a monom er in human body fluids, the three-dim ensional structure o f the dim er was n o t ideal fo r studies o f the in h ib ito ry site o f the monomer.
B u t the knowledge o f the structure o f the dim er allow ed us to construct a cystatin C variant w hich, by in tro d u ctio n o f an internal disulphide bond, was stable as a m onom er also when it was crystallized. The three-dim en
sional structure o f this cystatin C m onom er allowed firm conclusions about n o t o nly the in h ib ito ry site o f the native m onom eric protein, but also about other areas o f the molecule o f im portance fo r its stability.
The am yloid deposits o f the brain in A lzheim er’s disease and at o ld age consist m ainly o f am yloid [1-protein 42, a cleavage product o f the bigger amy
lo id precursor protein (APP), but significant amounts o f native cystatin C are co-deposited. This indicates that native cystatin C is prone to aggregate, which agrees w ith its tendency to dim erize du rin g crystallization. In the 1930s, an Icelandic physician A rn i Arnason noted that in some big families up to about 25% o f the members died o f cerebral haemorrhage before 30 years o f age.
A bout 40 years later, Gunnar Gudmundsson noted that it was am yloid depos
its w hich caused the brain haemorrhages in this disorder. B ut the chemical nature o f the deposits could n o t be determ ined. In 1983, a small am ount o f the protein constituting the am yloid deposits could be extracted and a short string o f its amino acid sequence determined. This sequence was identical w ith a part o f the sequence fo r cystatin C, w hich we previously had determ ined, and im m unohistochem ical studies confirm ed that the protein, indeed, was cystatin C. Further sequencing o f the extracted protein indicated that there was one amino acid difference, L6 8 Q , compared to the one fo r cystatin C determ ined earlier. In order to decide w hether this represented a disease- -causing m utation o r just a polym orphism , we produced a simple PCR-based
60
restriction fragm ent length polym orphism analysis to id e n tify the tw o gene variants (alleles). This was easy to do, since we already knew the nucleic acid sequences o f the m R N A and gene fo r cystatin C. The L68Q -m utation id e n ti
fied all the individuals w ith cerebral haemorrhage in the afflicted families. So, after about 50 years o f chemical progress, the firs t goal pinpointed by A rn i Arnason was fu lfille d : 1 m icro lite r o f blood allows to id e n tify the carriers o f the brain haemorrhage-producing gene. This has had a great effect fo r the clinical handling o f these patients. I t allows, fo r example, prenatal diagnosis.
B ut the second goal set up by A rn i Arnason, to elucidate the pathophysiology o f the disease so that it can be treated, is s till n o t reached.
T his second goal is m ore im p o rta n t than ju s t a llo w in g developm ent o f treatm ent options fo r the relatively few Icelandic patients carrying the L68Q -m utation, because the aggregation process, generating am yloid fib rils , m ig h t be operative also in m uch m ore com m on am yloid disorders, fo r instance fam ilial am yloid polyneuropathy as w e ll as Alzheim er’s and Parkin
son’s diseases. Knowledge o f the three-dim ensional structures o f m onom eric and dim eric cystatin C not only supported the hypothesis o f domain swapping as an im p o rta n t general mechanism in am yloid form ation, but also allowed id e n tifica tio n o f im portant cystatin C oligom ers preceding the form ation o f am yloid fib rils. Further studies using molecular m odelling, advanced physico
chemical methods, like atom ic force microscopy, sm all angle neutron scat
te rin g , nuclear magnetic resonance diffusom etry, production o f cystatin C variants w ith increased o r decreased tendency to oligom erize and synthesis o f peptidyl derivatives interfering w ith cystatin C oligom erization, have allowed detailed elucidation o f the aggregation process and development o f reagents delaying o r stopping it. This progress w ould n o t have been possible w ith o u t the knowledge and expertise o f Maciej Kozak, Sylwia Rodziewicz-M otowidło, Aneta Szymańska, Robert Kołodziejczyk, and other Polish scientists in Gdańsk and Poznań.
Even i f you have a specific plan fo r your research, fo r example elucida
tio n o f the structure and function o f cystatin C, you should always be open to your results being also o f interest to other areas o f science. For example, when we hypothesised that cystatin C was catabolised by the kidneys by being excreted by glom erular filtra tio n and subsequent degradation in the tubular cells, we studied patients w ith decreased kidney fu n ctio n (decreased glom er
ular filtra tio n rate). The results showed a strong inverse correlation between the cystatin C plasma level and the glom erular filtra tio n rate, and did not only support ou r hypothesis, but also indicated that the plasma, o r serum, level o f cystatin C is the best marker fo r kidney function. A fte r the firs t publication o f this in 1979, more than 3,500 studies have been published emphasizing the role o f cystatin C in m o n ito rin g kidney function. In some o f these studies,
61
Polish scientists, among them Joanna Poliak from C ollegium M edicum o f the Nicolaus Copernicus U niversity in Bydgoszcz, have played im p o rta n t roles.
Knowledge o f the structure and large size o f cystatin C has recendy allowed discovery o f a new renal disorder, designated “ Shrunken Pore Syndrome” , in w hich the glom erular pores have lost th e ir capacity to filte r big molecules like cystatin C, w h ile retaining th e ir capacity to filte r sm all molecules like creatinine. Shrunken Pore Syndrome has turned out to be a common disorder w ith a rem arkably hig h m ortality. Fortunately, the pathophysiology o f the disorder seems to allow treatm ent.
This longstanding and successful Polish-Swedish cooperation, in vo lvin g many m ore Polish scientists than those few m entioned in this account, and reported in more than 35 scientific articles, w ould probably n o t have been possible unless Zbigniew Grzonka had initiated regular Polish-Swedish meet
ings concerning the topics o f this discourse. These meetings started in Gdansk in 1996 and are ongoing w ith 14 meetings organised so far w ith the meeting venues alternating between Poland and Sweden.