10 Example for Fermentation in Complex (Liquid) Medium: Corallo pyronin A Producer Strain Corallococcus coralloides B035 Fermented in MD1 + G [4], A Standard Medium for Cultivation of Myxobacteria [7]

Compound Fermentation



4



59



Notes

1. Mind the minimum load of your balance and work with higher

concentrated stock solutions from which you can take an aliquot when you are dealing with very small sample weights.

2. In media that contains higher contents of sugars you should

autoclave the latter separately (concentration of the sugar solution max. 25 %) or rather sterile filtrate them into the media (to

avoid the formation of growth inhibiting caramelization and

Maillard reaction byproducts). Particular high concentrated

sugar solutions (double-strength concentrates) or solutions

with unsolved sugar and peptone sediments, as well as media

that contain high amounts of phosphate salts as well as media

with alkaline pH above 8.0, are affected.

3. Metal ions like Ca2+, Mg2+, Fe2+/3+ may precipitate during autoclavation, especially in solutions with higher pH values and in the

presence of higher phosphate amounts. This generally does not

inhibit microbial growth. Therefore, direct addition to a medium

before autoclavation can be tolerated. For more accurate condition and especially if you compare different media conditions the

addition in the form of a sterile filtered solution is advised.

4. A fast check for the right temperature is to hold the autoclaved

glass bottle against your cheek. If it is just warm and not painful, thermolabile ingredients can be added.

5. Simple Bubble point test to check for filter integrity: Release

the filter from the syringe; Draw 10 mL of air into the syringe;

Reattach the wetted filter and cannula to the syringe; Dip the

tip of the prepared syringe into a water-filled vessel; Slowly

push in the plunger until the first air bubbles are released; If air

bubbles are released before the air is compressed to 2 mL filter

integrity is doubtful and has to be repeated. This is a quick and

easy but rough approximation. A more accurate alternative is

the utilization of a manometer that allows determining an

accurate value for the bubble point that can be compared with

the sterile filter bubble point value given by the manufacturer.

6. For solid media with pH values < 5.0 higher agar concentrations are required for sufficient gel stability.

7. Dilution plating: Work in a laminar airflow system; Sterilize an

inoculation loop in the external tap of a Bunsen burner; Let

the inoculation loop cool down for 15 s; Check if the inoculation loop has sufficiently cool downed by tapping the inoculation loop on unused spots at the corner of the agar plate; Pick

a single microbial colony from a petri dish or tap the inoculation loop in a microbial liquid suspension; Start at the left

upper side of the plate and streak out serpentine like several



60



Henrik Harms et al.



times to the opposite right side, slide over the agar and do not

penetrate the agar surface; Sterilize the inoculation loop in the

external tap of a Bunsen burner again; Turn the plate around

90° streak out to the other side, starting at the end of the last

streak; Sterilize the inoculation loop in the external tap of a

Bunsen burner again; Turn the plate around 90°, streak out to

the other side, starting at the end of the last streak.

8. Often the desired compound is only produced under a special

growth condition. Therefore, it is of advantage to test samples

from different fermentation time points for the yield of the

compound. In that way the ideal conditions, e.g., how long

fermentation should be, can be determined.

9. For many bacteria it is known that antimicrobial compounds

are not produced in complex media. Instead, a certain shortage in the fermentation broth, e.g., low phosphate availability,

can trigger the biosynthesis. Thus, the use of different media

for the incubation of the preculture, and of the main culture

might be of advantage. A complex medium, in which the

microorganism growth is fast, can be used to produce enough

cell material for the preculture. This forms the inoculum for

the main culture which is fermented in the optimized medium

for production.

10. To get more insights into the critical parameters for production,

bioreactors instead of shaking flask experiments can be used.

By monitoring and/or regulating parameters in the fermentation broth, e.g., pH and O 2-saturation, the yield of the

fermentation is increased.

11. The supplementation of the fermentation with substances that

trigger the production might be advantageous. Although mostly

the regulation mechanisms of the biosynthesis are not known,

e.g., factors that trigger production positively, some compounds

like hormaomycin [8] have been used to induce production.

12. It might be even the case that the production of a compound

is only triggered by the presence of another microorganism.

Thus, it can be that the idea of an axenic culture for fermentation is not always applicable.

13. For bacteria that particularly rely on interspecies messenger

substances for their development, e.g., some marine myxobacteria, it can be beneficial not to add adsorber resins like

Amberlite XAD 16 or Sephabeads directly, but rather a few

days after the incubation has started.

14. Sedimentation of cells can be significantly enhanced by

centrifugation of the cryo tubes for 5 min at 4000–6000 × g

for bacteria and 2000–3000 × g for yeasts.



Compound Fermentation



61



15. The cryo tube can be placed in a Styrofoam box (each cryo

tube should be surrounded by 10–15 mm isolation material);

Optimal is a low cooling rate about −1 °C/min to −30 °C, followed by a fast cooling rate to −70 °C, since this prevents the

formation of intracellular ice crystals; For the reactivation of

the cryo cultures the frozen cryo tube solution should be

quickly warmed up in a tempered water bath (mostly 37 °C,

but this has to be in accordance with the growth conditions of

the microorganism), since this prevents intracellular recrystallization to larger crystals during thawing.



Acknowledgment

This work was supported by the German Federal Ministry of

Education and Research (BMBF) through the German Centre for

Infection Research (DZIF) initiative.

References

1. Balows A, Trüper HG, Dworkin M et al (1992)

The Prokaryotes. Springer, New York, NY

2. Atlas RM (2010) Handbook of microbiological media, 4th edn. ASM Press ; CRC Press/

Taylor & Francis, Washington, D.C. : Boca

Raton, FL

3. Harms H, Kehraus S, Mosaferan DN et al

(2015) Aβ-42 lowering agents from the marinederived fungus Dichotomomyces cejpii.

Steroids 104:182–188

4. Erol O, Schäberle TF, Schmitz A et al (2010)

Biosynthesis of the myxobacterial antibiotic

corallopyronin A. Chembiochem 11(9):

1253–1265



5. Garrity G (ed) (2001–2011) Bergey’s manual

of systematic bacteriology, 2nd edn, vol 1–5.

Springer, New York, Berlin, Heidelberg etc

6. Koser SA (1968) Vitamin requirements of

bacteria and yeasts. Charles C. Thomas

Publisher, Springfield, IL

7. Behrens J, Flossdorf J, Reichenbach H (1976)

Base composition of deoxyribonucleic acid

from Nannocystis exedens (Myxobacterales). Int

J Syst Bacteriol 26(4):561–562

8. Höfer I, Crüsemann M, Radzom M et al (2011)

Insights into the biosynthesis of hormaomycin,

an exceptionally complex bacterial signaling

metabolite. Chem Biol 18:381–391



Chapter 4

Structure Elucidation of Antibiotics by NMR Spectroscopy

Georgios Daletos, Elena Ancheeva, Raha S. Orfali, Victor Wray,

and Peter Proksch

Abstract

Nuclear magnetic resonance (NMR) spectroscopy is a powerful tool for the structure elucidation of

antibiotics in solution. Over the past 30 years there have been numerous publications describing the use

of NMR to characterize naturally derived or synthetic antibiotics. A large number of one-dimensional (1D)

and two-dimensional (2D) NMR methods are available today and the list continues to expand. In this

chapter, we will consider the key NMR experiments that provide useful information for compound structure elucidation.

Key words NMR, Structure elucidation, Antibiotics, Natural products, Callyaerin A



1



Introduction

Natural products and their derivatives have historically been an

untapped source of antibacterial leads used in the development of

drugs to treat bacterial infections. Examples of naturally derived

antimicrobial agents of different antibiotic classes are penicillin G

originally isolated from Penicillium notatum, erythromycin from

Saccharopolyspora erythraea, streptomycin from Streptomyces griseus, and polymyxins from Bacillus polymixa, among others. Most

of these compounds were developed by screening soil-dwelling

microorganisms, such as actinomycetes, during the golden era of

antibiotic discovery in the 1940s to 1960s [1]. In recent years,

however, microbial pathogens are becoming increasingly resistant

to clinically applied antibiotics and pose a global threat to human

health [2]. Thus, there is an overwhelming need to search for new

antibiotics with new targets and novel mechanisms of action.

Nature has inspired chemists and biologists alike for many

decades, providing a rich and unprecedented diversity of evolutionary preselected lead structures unparalleled even by the largest

combinatorial databases [3]. Since there are still many unexplored

resources in nature, the potential for finding new bioactive



Peter Sass (ed.), Antibiotics: Methods and Protocols, Methods in Molecular Biology, vol. 1520,

DOI 10.1007/978-1-4939-6634-9_4, © Springer Science+Business Media New York 2017



63



64



Georgios Daletos et al.



compounds that could be optimized to yield therapeutic agents is

also enormous. This fact coupled with advances in approaches for

natural-product isolation and identification could open the door to

a new era in the investigation of antibiotics [4, 5].

As part of our ongoing search for new antibiotic natural products, chemical investigation of the Indonesian sponge Callyspongia

aerizusa revealed a group of unusual cyclic peptides, called callyaerins [6, 7]. The basic structural unit of the callyaerins comprises

a cyclic peptide with a linear peptide side chain, both of variable

size, linked through a nonproteinogenic (Z)-2,3-diaminoacrylic

acid (DAA) moiety. Among the isolated peptides, callyaerin A

showed potent activity against Mycobacterium tuberculosis with

MIC90 value of 2 μM (Fig. 1). In addition, callyaerin A exhibited

no cytotoxicity toward THP-1 (human acute monocytic leukemia)

or MRC-5 (human fetal lung fibroblast) cells (IC50 > 10 μM),

which highlights the potential of this compound as a promising

lead for new antitubercular agents [7].

The structure elucidation of callyaerins was a challenging task

and was based mainly on extensive nuclear magnetic resonance

(NMR) spectroscopic analysis. NMR spectroscopy is an extremely

powerful and widely used method for the structure determination

of natural products in solution, including antibiotics [5, 8]. NMR

experiments are based on the magnetic properties of atomic nuclei.

When placed in a powerful homogeneous magnetic field, certain

nuclei (e.g., 1H, 13C, 15N) undergo resonance at specific radio frequencies in the electromagnetic spectrum to produce signals that

can be readily detected. Detailed analysis of these diagnostic signals

in the NMR spectrum provides unique information about the

molecular structure of the compound under investigation. Over

the past 30 years, major improvements in spectroscopic instrumentation hardware allow structure analysis to be carried out on submilligram amounts of a compound [9]. Moreover, the advent of

multidimensional NMR techniques has revolutionized structure

elucidation of natural products so that, in the large majority of

cases, unambiguous structural information is obtained, even for



Fig. 1 (a) Structure of callyaerin A. (b) Antitubercular (Mtb) and cytotoxicity (THP-1 and MRC-5) profile of

callyaerin A



Nmr Analysis of Antibiotics



65



highly complex molecular structures [10]. Herein, we offer an

overview of the one-dimensional (1H, 13C, DEPT) and twodimensional (COSY, TOCSY, HSQC, HMBC, and NOESY/

ROESY) techniques that were successfully applied in the case of

callyaerins and have proved most useful for the structure elucidation of other naturally derived or synthetic antibiotics.



2



Materials

1. Glass vial containing freeze-dried, purified compound.

2. NMR tube (see Notes 1 and 2).

3. Deuterated solvent, such as CDCl3, DMSO-d6, or MeOH-d4

(see Notes 3 and 4).

4. Glass Pasteur pipettes, natural rubber teats, glass wool, and

Parafilm® (see Note 5).



3



Methods



3.1 Sample

Preparation



1. Dry the purified compound (see Note 6).

2. Use a glass Pasteur pipette for adding the suitable deuterated

solvent to the sample. (see Note 7).

3. If necessary filter your sample using glass wool (see Note 8).

4. Pipette the sample into an NMR tube (see Notes 9 and 10).

5. Wrap a 1 cm ì 2 cm Parafilmđ strip around the cap of the NMR

tube (see Note 11).

6. Label your sample at the top of the NMR tube.



3.2 Useful

Parameters from NMR

Measurements

3.2.1 Chemical Shift



At this point, it is important to emphasize the most basic parameters measured in the NMR spectra, including the chemical shift,

coupling constant, and integration.

1. The chemical shift (δ) is a measure of the resonant frequency

of an NMR-active nucleus (e.g., 1H, 13C, or 15N) and is quoted

in parts per million (ppm) (see Note 12).

2. The chemical shift denotes the positions of the NMR peaks

relative to a reference compound (usually tetramethylsilane,

TMS) (see Note 13).

3. By convention the shielded signals of TMS are set to 0 ppm,

situated on the right side of the chemical shift scale in the

NMR spectrum. The resonances of common functional groups

are less shielded, that is they have higher shifts, and are placed

on the left side of the signal of TMS [11, 12].



Georgios Daletos et al.



66



3.2.2 Coupling Constant



1. Coupling constant is the absolute separation between two or

more peaks (splitting) of each NMR signal, arising from coupling between nuclei, such as proton-proton or proton-carbon.

2. This intramolecular communication, caused by through-bond

interactions of nuclei, is the phenomenon known as spin-spin,

scalar, or J-coupling (see Note 14).

3. J-couplings are measured in cycles per second (Hz) and their

magnitude depends on their distance apart, relative stereochemistry, and chemical environment.

4. The observation of J couplings is important, as useful structural information can be derived, including bond linkage (see

Note 15) and molecular conformation (see Note 16).



3.2.3 Integration



1. In the 1H NMR spectrum, integral values of peak areas underneath the NMR signals are proportional to the number of

hydrogen atoms of related functional groups in the sample

(see Note 17).

2. In sample mixtures the quantitative relationship between the

corresponding components can be determined by the different

ratio of the integrals in the spectrum.



3.3



3.3.1



1D NMR Methods



1



H NMR



The 1D NMR spectrum is a plot showing amplitude along a frequency axis, which is typically the chemical shift axis. To obtain this

spectrum, the nuclei are irradiated and generate a signal that is

detected in the time domain and then converted mathematically

into the frequency domain by employing a mathematical procedure known as Fourier transformation [11].

1. By measuring a 1H NMR spectrum, we observe frequency

ranges of 1H resonances with common chemical shifts from 0

to 12 ppm. Thus, a typical spectral window for 1H NMR is at

least 12 ppm wide (see Note 18). An example of the 1H NMR

spectrum of callyaerin A in DMSO-d6 is shown in Fig. 2. 1H

NMR analysis is performed as follows.

2. Standardize the reported chemical shifts with reference to the

residual solvent peak.

3. Integrate the 1H NMR spectrum to obtain a total hydrogen

count (see Note 19).

4. List all 1H NMR chemical shifts to two decimal places (see

Note 20).

5. List the multiplicities and coupling constants (J in Hz) for all

1

H NMR signals (see Note 21).

6. Inspect the spectrum and identify obvious functionalities, such

as aromatic protons (∼ 7.0–8.0 ppm), methoxyl (∼ 3.5–4.0

ppm) or methyl (∼ 1.0–2.0 ppm) groups, from their characteristic shifts, multiplicities, and integrations (see Note 22).



Nmr Analysis of Antibiotics



67



Fig. 2 1H NMR (600 MHz, DMSO-d6) spectrum of callyaerin A



3.3.2



13



C NMR



1. The 13C NMR spectrum shows the chemical shifts of carbon

resonances, and thus the total carbon atom number of a molecule. The frequency range for common 13C shifts is from 0 to

220 ppm. Due to the low isotopic abundance (1.1 %) of 13C, as

well as its inherent low sensitivity (~1/64 to that of 1H), the

signals are weaker than those of 1H, and thus more time for

spectra recording is acquired (see Notes 23 and 24). The 13C

NMR spectrum of callyaerin A is shown in Fig. 3. 13C NMR

analysis is performed as follows.

2. Reference the

solvent peak.



13



C NMR spectrum to the residual deuterated



3. List all 13C NMR chemical shifts to one decimal place

(see Note 25).

4. As in the case of the 1H spectrum, inspect the 13C spectrum for

obvious functionalities, such as carbonyl (∼ 170–220 ppm),

aromatic (∼ 110–130 ppm), methoxy (∼ 50–60 ppm), or

methyl (∼ 10–30 ppm) groups. [13].

3.3.3 DEPT



1. Distortionless enhancement by polarization transfer (DEPT) is

the usual method for determining the type of carbon atoms

present: tertiary (CH), secondary (CH2), and primary (CH3)

carbons (see Note 26).



68



Georgios Daletos et al.



DMSO



180



170



160



150



140



130



120



110



100



90

80

f1 (ppm)



70



60



50



40



30



20



10



0



Fig. 3 13C NMR (150 MHz, DMSO-d6) spectrum of callyaerin A



2. The DEPT-135 spectrum displays CH’s and CH3’s as positive

singlet peaks and CH2’s as negative singlet peaks, whereas the

quaternary carbons are absent. The DEPT spectrum of callyaerin A is shown in Fig. 4. Interpretation of the DEPT spectrum is performed as follows.

3. Determine the carbon atom order according to the phase of

the peaks and in the case of quaternary carbons by direct comparison with the 13C NMR spectrum.

4. Count the number of hydrogen atoms (H’s) that are directly

attached to the respective carbon atoms (13C’s).

3.4 2D-NMR

Methods



1. A 2D NMR spectrum is obtained using multipulse experiments that correlate signals from two frequency domains (f1

and f2). Contour plots show cross peaks that associate information on one axis with information on the second axis [11].

2. These methods are valuable tools for structure determination

of complex compounds, since either through-bond or throughspace interactions are revealed between nuclei of the same

(homonuclear—typically proton) or different (heteronuclear—

typically proton and carbon) elements (see Note 27).

3. In addition, cross peaks observed in 2 D spectra allow the

assessment of accurate chemical shift values and J-couplings

that cannot be assigned directly from the 1 D spectrum due to

signal overlap.



3.4.1 COSY



1. The homonuclear shift correlation spectroscopy (1H, 1H-COSY,

or COSY) spectrum shows the through-bond coupling

connectivities between groups containing hydrogen atoms,



69



Nmr Analysis of Antibiotics



CH, CH3



DMSO



CH2

180



170



160



150



140



130



120



110



100



90

80

f1 (ppm)



70



60



50



40



30



20



10



0



Fig. 4 DEPT-135 NMR (75 MHz, DMSO-d6 ) spectrum of callyaerin A



based on geminal (2J) and vicinal (3J) proton couplings (see

Notes 28 and 29).

2. The 1H spectrum of the sample is set on both horizontal (f2)

and vertical (f1) axes. Autocorrelated peaks appear on the diagonal (δ1 = δ2), which is the symmetrical axis of the COSY

spectrum.

3. A signal situated off the diagonal is called a cross peak and

appears whenever protons with resonances at δ1 and δ2

(δ1 ≠ δ2) are coupled to one another.

4. A pair of coupled protons can be identified by lines through

the cross peak, which is symmetrical with respect to the diagonal (see Note 30). An example of the COSY spectrum of callyaerin A is shown in Fig. 5. Analysis of the COSY spectrum is

performed as follows.

5. Draw a vertical line from a known diagonal peak (HA) until

you connect with a cross peak (HA, HB). The horizontal line

from this cross peak to the diagonal identifies the shift of the

coupled proton (HB).

6. In a similar manner, projecting from the last diagonal peak to

the next cross peak and then back to the diagonal allows the

assignment of the whole coupling network in the molecule.



70



Georgios Daletos et al.

Val-H3(g ), H3(g ′)

Val-Ha Val-Hb

Val-NH



H

N



CH3 (g )



0



CHb CH3 (g ¢)



1



CHa C



2

O



4

5



f1 (ppm)



3



Val-Ha



6

7

8

9



Val-NH

9.5 9.0 8.5 8.0 7.5 7.0 6.5



6.0 5.5 5.0 4.5



4.0 3.5 3.0 2.5 2.0



1.5 1.0 0.5 0.0 -0.5



f2 (ppm)



Fig. 5 COSY NMR spectrum of callyaerin A (600 MHz, DMSO-d6)



The coupling network of the valine residue of callyaerin A is

shown in Fig. 5.

7. Check the 1H NMR spectrum to confirm that J couplings and

integrals are in agreement with the assignments made by

COSY.

3.4.2 TOCSY



1. The total correlation spectroscopy (TOCSY) experiment allows

the generation of cross peaks between virtually all protons

within a given spin system (see Notes 31 and 32).

2. The 1H spectrum of the sample is set on both horizontal (f2) and

vertical (f1) axes and cross peaks are situated off the diagonal line

(see Note 33). The TOCSY spectrum of callyaerin A is shown in

Fig. 6. Analysis of the TOCSY spectrum is performed as follows.

3. Draw a vertical line through a selected peak at the top of the

TOCSY spectrum. This line will pass through one or more

cross-peaks, which are situated off the diagonal (see Note 34).

4. From these cross-peaks project horizontal lines to the 1H spectrum set along the vertical axis. The respective peaks allow the

detection and assignment of all mutually coupled proton signals within a specific spin system (see Notes 35).