Remembering Malcolm J. Casadaban

Abstract

Malcolm J. Casadaban died on 13 September 2009 from an infection and was found to have a weakened strain of the bacterium Yersinia pestis in his blood. This tragic event took the life of one of the most creative and influential geneticists of our time. In the late 1970s and '80s, Malcolm invented novel approaches which changed the way many of us did science. Jon Beckwith, Tom Silhavy, and Olaf Schneewind have chronicled his scientific life from graduate school to his death and give us insight into Malcolm's genius.

Philip Matsumura Editor in Chief, Journal of Bacteriology

IN MEMORIAM

Malcolm J. Casadaban, who passed away on 13 September 2009, was an imaginative experimentalist whose technological and intellectual innovations were used by a generation of scientists employing genetic approaches in microbes. Malcolm's methods even transcended the typical divide that separates investigators of microbial and higher eukaryotic systems.

Born in New Orleans in 1949, Malcolm was the first of seven children raised by John and Dolores Casadaban. After earning his diploma from Jesuit High School in his home town, he studied biology at the Massachusetts Institute of Technology (MIT) and graduated in 1971. Inspired by his teachers at MIT, Malcolm pursued graduate training at Harvard Medical School, where he joined the laboratory of one of us (J.B.) and completed his Ph.D. requirements in 1976. Even though he was only at the beginning of his career, this gentle, naïve young man revolutionized genetic approaches to studying a host of biological problems. Starting in the early 1970s, in a series of papers from the Beckwith laboratory (4-6) and subsequently with his postdoctoral fellowship mentor (10, 11), Stanley Cohen, at Stanford University, Malcolm presented to the biological world increasingly sophisticated methods for construction of gene fusions in vivo. Even though the recombinant-DNA era later brought in a new in vitro technology that allowed construction of fusions to genes individually, Malcolm's approach still had broad uses. Importantly, it allowed scanning of chromosomes by using gene fusions, a technique that has been applied to many problems (37).

Malcolm's graduate mentor (J.B.) recalls his unusual personality. He loved to construct bacterial strains with new properties, bringing together complex, almost architectural plans. Malcolm once explained that his interest in genetic constructions originated from his youth, as he grew up watching his father work as a carpenter. Malcolm's strain constructions sometimes involved steps that went against tradition, for instance, mixing together transduction and conjugation steps in one tube. Tracing the origins of widely used Casadaban strains such as MC1000 and MC4100 can be a rapidly confusing enterprise because he involved 20 or more precursor strains (6). These strains also illustrate Malcolm's ingenuity and the nature of his work products, which were so clearly his own. For these reasons, Jon Beckwith told Malcolm that he should publish the papers reporting his graduate work on gene fusion techniques under his name alone (4-6).

Gene fusion is used so commonly now that the origins of this powerful technology have become obscure. Before Malcolm, the number of existing gene fusions could be counted on the fingers of one hand. They were constructed by giants in the field of bacterial genetics using cumbersome multistep procedures that were difficult to generalize (2, 24). Malcolm devised a three-step procedure for constructing gene fusions that could be generalized for any nonessential gene in Escherichia coli. At the heart of Malcolm's method is a plaque-forming λ lac-specialized transducing phage called λp1(209) (6). It is important to note that this phage was constructed with toothpicks (36); no recombinant-DNA methodology was involved. To appreciate Malcolm's genius, the reader is encouraged to look at the paper describing this phage construction (6).

The first step in Malcolm's procedure was to isolate a Mu insertion in the gene of interest in a strain like MC4100, which carries a deletion of the lac operon (6). This strain is then lysogenized with λp1(209). Since λp1(209) lacks an attachment site, it can integrate into the chromosome only by homologous recombination between the Mu DNA elements, and this recombination places the lac operon within the gene of interest. Provided that the Mu prophage is integrated in the proper orientation, it also places the lac operon downstream from the promoter of the gene of interest. Since the Mu prophage carries a temperature-sensitive mutation in the repressor gene, selection for the Lac⁺ phenotype at high temperatures will yield survivors that have suffered a deletion event that removes the lethal Mu prophage and fuses the lac operon to the promoter of the gene of interest (6). Fusions isolated by this method are called operon or transcriptional fusions, and researchers in many labs took advantage of Malcolm's method to study their particular gene or regulatory system of interest (37). For example, malP-lacZ⁺ fusions were used to isolate the first constitutive mutations in the regulatory gene malT, and these mutations were used to show that MalT regulates the mal operons in a strictly positive fashion (15). Without fusions, there was no way to isolate such constitutive mutations because chemists had not been able to synthesize noninducing substrates of the mal regulon. With fusions, lactose becomes a noninducing substrate for any regulatory system. Simply selecting for the Lac⁺ phenotype with these fusion strains yields constitutive mutations.

Malcolm cleverly modified his three-step method so that it could be used to isolate true gene fusions, i.e., fusions that create a hybrid gene that specifies a hybrid protein with functional LacZ at the carboxy terminus. He did this by introducing a nonsense mutation at codon 18 of lacZ, producing the specialized transducing phage λp1(209,118) (6). Of course, this was also done with toothpicks. Now to get Lac⁺ survivors at high temperatures, the deletion that removes the lethal Mu prophage must also enter the lacZ gene and remove the nonsense mutation. The target for the end point of this deletion is very small; the deletion must remove the nonsense mutation, but it can't extend much further into lacZ without destroying LacZ function. Accordingly, survivors are rare, but these fusion strains proved to be of tremendous value. For example, they opened the door for genetic analysis of protein secretion. When the amino terminus of a periplasmic protein such as MalE or an outer membrane protein such as LamB is fused to LacZ, the cell attempts to secrete the hybrid protein (1) (38). It turns out that LacZ doesn't fold properly if secreted from the cytoplasm, and the unexpected Lac⁻ phenotype of such fusion strains allowed the isolation of the first signal sequence mutations (18) and the identification of genes that specify components of the cellular protein secretion machinery (17, 31).

There are many pitfalls in Malcolm's three-step method, and consequently, many who wanted to use gene fusion technology were reluctant to try it. So as a postdoctoral fellow in Stanley Cohen's laboratory, Malcolm developed a one-step method (11). For this method, Malcolm inserted a promoterless lac operon, together with the gene for β-lactamase (penicillin resistance), within the Mu genome to produce a defective specialized transducing phage, Mud(Ap, lac). Malcolm used recombinant DNA in the construction of this phage, but only at one of many steps. One of the intermediates in this construction would be very rare and extremely difficult to identify. In typical Malcolm fashion, he never did identify it. He simply assumed it must be there and carried on. Clearly, this intermediate must have been there! His report is another paper that provides real insight into his genius (11). In a subsequent paper, Malcolm and his wife, Joany Chou, describe a related Mud(Ap, lac) phage that can be used to construct fusions that make LacZ hybrid proteins (7).

The Mud(Ap, lac) phages were the first of the fusion-generating transposons. With these elements, constructing fusions is really simple (7, 11). You infect a desired strain with the Mud(Ap, lac) phage, for example, wait 20 min, and then plate the infected cells onto medium that selects for the loss of a gene of interest and contains ampicillin and a LacZ color indicator such as X-Gal (5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside). Then you go have a beer, watch a football game, and come back in the morning to purify your fusions (the blue colonies). It is true that modern technologies such as PCR have greatly simplified the construction of gene fusions to known genes. However, fusion-generating transposons have additional uses. For example, they can be used to find unknown genes that belong to a particular regulon. This is done simply by looking for fusions that are regulated in a particular way. In this manner, the genes that specify the proteins that comprise the redundant arabinose transport systems were identified (26). This could not be done in traditional ways because mutants lacking only one of the transport systems have no relevant phenotype. Similarly, the transposons were used to identify unknown genes regulated by the SOS response (25). Nowadays, fusions are commonly employed and there are many reporters other than LacZ, with green fluorescent protein (GFP), the discovery of which led to a Nobel Prize, being just one example (13). But we should not forget that it was Malcolm's work that popularized this experimental approach and first made it generally available.

In 1980, Malcolm accepted a faculty appointment at the University of Chicago. One of many immediate research foci was the characterization of E. coli Tn3 transposase and its regulation (9). Once again, Malcolm was aiming to develop facile tools for transposon mutagenesis (8, 14, 16). Other work expanded the uses of gene fusion in E. coli and examined the universality of this technology for other systems, a purpose that led Malcolm to express β-galactosidase in yeast (12, 35). Malcolm's first graduate student, Alfonso Martinez-Arias, used lacZ fusions to isolate the upstream regulatory element of the yeast leu2 gene, which enabled molecular studies on transcriptional repression by leucine at this promoter (27, 28). In a collaboration between Malcolm's and Donald Steiner's laboratories, Malcolm's lacZ fusion technologies enabled the first study of insulin gene expression in pancreatic β-cells (29, 30).

Genetic technology was developed further by his graduate student Eduardo Groisman, who created mini-Mu phages carrying a plasmid replicon and antibiotic resistance cassettes for in vivo cloning (22, 23). The power of this technology was revealed through the rapid isolation of mutants with mini-Mu transposon insertions and measurements of gene expression via lacZYA fusion. Desired mutations could then be studied by transducing loci into other genetic backgrounds by using Mu helper phages or rapid recombination cloning of gene fusions via the associated plasmid replicons. This technology was expanded for λ helper phages and cosmids, providing a suite of tools for rapid generation, isolation, and transfer of mutations in E. coli and other Gram-negative bacteria (20, 21).

By 1990, Malcolm directed his research to address the need for genetic tools for Gram-negative pathogens, initially studying Pseudomonas aeruginosa and its transposable bacteriophages and then Salmonella enterica serovar Typhimurium, in which he identified core lipopolysaccharide genes to demonstrate the use of his tools (32-34). At the same time, Malcolm was engaged in founding and developing a biotechnology company, ThermoGen Inc., exploring the genes of the archaeon Thermus flavus for thermostable enzymes with industrial uses (40, 41). This work was carried forward by David C. Demirjian, Malcolm's last Ph.D. student.

In pursuit of new research programs, Malcolm collaborated with one of us (O.S.) on the type III secretion pathway of pathogenic Yersinia species. Malcolm initially developed tools for transposon mutagenesis as well as gene transfer and mapping. He eventually set out to identify mutations that affect the low-calcium responses of these microbes. Briefly, chelation of calcium ions activates type III secretion, and this arrests the growth of yersiniae (19). Translational hybrids of secretion substrate genes were generated as lacZ fusions, and the resulting proteins can block type III secretion, enabling the selection of mutations that abrogate substrate recognition and suppression of those mutations (3, 39). In the midst of these experiments, Malcolm unexpectedly passed away, his last studies still unfinished.

Already as a graduate student, Malcolm was an endearingly, and sometimes frustratingly, naïve person. He might absorb and mention something about some news item or some political perspective, but it was not something that he was going to put a lot of thought into. His mind was fully involved with his science, his constructions. Malcolm was very soft-spoken, with maybe the remnants of a Louisianan twang that made him occasionally hard to hear. Yet he loved to help people. In the laboratory, at conferences, or on the phone—he tutored the many people who were using his strains about their intricacies. He clearly loved to teach and was unstintingly ready to instruct, however long it took.

Malcolm J. Casadaban's ingenuity at inventing genetic tools and his dedication to experimental work on complex biological problems will continue to serve as an inspiration for the scientific community. He is survived by his parents, brothers, and sisters; by his two daughters, Brooke and Leigh, and his former wife, Joany Chou; and by his fiancée, Casia Holmgren.

Figure 1.

Malcolm J. Casadaban 1949-2009