Bacteriochlorins and their metal complexes as NIR-absorbing photosensitizers: properties, mechanisms, and applications

Bacterichlorins possess a characteristic intense electronic absorption in the near-infrared part of the elec- tromagnetic radiation (NIR) from 700 nm up to 900 nm, where endogenous chromophores do not absorb and which allows deep penetration through tissues. Although naturally-occurring metal complexes derived from bacteriochlorophylls are unstable and their use is limited, it is currently possible to obtain large-scale libraries of photostable synthetic compounds belonging to the bacteriochlorin family. This review presents an up-to-date overview of the most signiﬁcant studies on the synthesis, spectroscopic, photochemical, and electrochemical characteristics as well as various potential applications of bacteri- ochlorins and their metal complexes. Particular emphasis has been given to the possibilities of their use in medicine, especially in photodynamic therapy of cancer (PDT), photodiagnosis (PD), and photody- namic inactivation (PDI) of bacteria, viruses, and fungi. The combination of (metallo)bacteriochlorins with polymeric micelles, lipoproteins, nanoparticles, and metal-organic frameworks in order to increase their efﬁcacy is also discussed. Other potential applications of (metallo)bacteriochlorins discussed in this paper include their use as light-harvesting antennas, optical sensors, photocatalysts, and dye-sensitized solar cells for efﬁcient solar energy conversion. DSSC, dye-sensitized solar cells; EBRT, external beam radiotherapy; EMA, European Medicine Agency; HOMO, highest occupied molecular orbital; human serum albumin; internal conversion; IL-6, interleukin-6; ISC, intersystem crossing; chemokine keratinocyte chemoattractant; density index; occupied molecular adenine dinucleotide benzodiazepine perchlorate; possess lower ﬂuorescence quantum yield ( U F ) compared to their monomers. This study shows that the geometry and spectral properties of the investigated dyad are similar to those observed in particular pairs involved in photosynthetic ﬂuorescent labeling of biomolecules, photodiagnosis (PD), solar energy conversion, optical sensing, and photocatalysis. The recent results from various scientiﬁc groups, including our, have shown that bacteriochlorins are promising compounds for PDT because they can use NIR radiation that deeply penetrates through the tissues and convert it very effectively to highly reactive oxygen species such as singlet oxygen, superoxide ion, hydrogen peroxide and hydroxyl radicals. In vitro studies with several families of (metallo)bacteriochlorins generally revealed their low cytotoxicity in the dark and high phototoxicity to various cancer cells. In a few independent studies, it was demonstrated that the photodynamic effect does not depend on the quantum yields of singlet oxygen formation, but it is strictly correlated with the production of oxygen-centered radicals. Bacteriochlorins that generate free radicals more efﬁciently lead to a more effective photodynamic effect than photosensitizers generating only singlet oxygen. The experiments performed on different animal models showed favorable pharmacokinetics, biodistribution and photodynamic efﬁcacy. It was also demonstrated that the functionalization of bacteriochlorins may increase the effectiveness of the applied therapeutic protocols with polymeric micelles, lipoproteins, nanoparticles and metal-organic frameworks.


Introduction
The interaction of light with biologically active compounds plays a key role in the living organisms' function. Light energy is a part of electromagnetic radiation that can lead to heating, mechanical effects, and chemical reactions. Thus, the transfer of energy through photon absorption results in a number of processes and reactions essential for biology, industry, and medicine. Moreover, numerous medical strategies employ light to visualize the interior of the body and are fundamental to its further application in a variety of more sophisticated, modern diagnostic and therapeutic tools [1,2]. Therefore, the study of the light-driven processes addresses issues of the most importance for both scientific research and the quality of health care. The light absorption in tissues is strongly dependent on the absorbed wavelength (k max ). Ultraviolet (UV, 100-390 nm) absorbed by DNA, amino acids and proteins, is more energetic electromagnetic radiation than visible light (Vis, 390-720 nm) that, in turn, may be absorbed by endogenous pigments such as hemoglobin, vitamins and other biologically important molecules. Consequently, these forms of radiation do not pass through the body (UV) or do not penetrate deep enough into the tissues (Vis) [3]. On the other hand, near-infrared (NIR, 720-900 nm) is a low energy range of light where photons exhibit a maximum depth of penetration through tissues. It is also worth noting that the density of photon flux on the Earth's surface reaches a maximum at approximately 700 nm, and up to 40% of VIS + NIR photons reaching the Earth's surface are, in fact, NIR photons [4] (Fig. 1).
NIR photons are the most penetrating and least hazardous to human tissues and can, therefore, provide direct access to biological targets. The range of spectra between 650 and 850 nm, where light penetrates tissue up to 1-3 cm, has been termed as a phototherapeutic window (Fig. 1). It is a low energetic part of the spectrum, where the compounds present in living tissues do not absorb light, but it also corresponds to energies still sufficient for photochemical reactions (e.g., generation of singlet oxygen through direct energy transfer). Recent reports have demonstrated the broad applications of NIR photons in photomedicine to facilitate biostimulation [5], imaging [6,7], diagnosis [8], and therapy of many diseases [9]. NIR photons are also extensively used in optical communication and could potentially be the most energy-effective in photocatalytic processes [4,10]. Unfortunately, their potential has not been fully exploited for many years due to the lack of chromophores, which by absorbing NIR radiation, would simultaneously be sufficiently (photo)stable. Thus, the synthesis of new photo-and thermally-stable chromophores characterized by intense absorption in the NIR is essential for their implementation in solar energy conversion, photomedicine, sensing, and photocatalysis.
Tetrapyrroles are a class of macrocyclic compounds that are commonly found in nature and, by performing various biochemical functions, are essential for the proper functioning of the majority of living organisms [11][12][13]. They are present in many enzymes and proteins whose functions include metabolism, transport of molecular oxygen, as well as electron transfer and energy conversion processes [11,13,14]. Moreover, they can be used as effective catalysts for numerous chemical and photochemical reactions [15,16]. Tetrapyrroles are classified into three main classes: porphyrins, chlorins and bacteriochlorins. An example of a porphyrin-based biologically active compound is heme present in hemoglobin (Fig. 2, blue). The most studied chlorin-type molecule is chlorophyll a (Chl a) responsible for the green color of plants, algae, and cyanobacteria (Fig. 2, green). Whereas bacteriochlorophyll a (BChl a), involved in the photosynthesis in some bacteria, belongs to the family of bacteriochlorins (Fig. 2, red). Although at first glance, the difference between these macrocycles seems to be insignificant, their spectroscopic, photophysical and redox properties differ considerably, which will be discussed in detail in this review. Unlike porphyrins, which possess a full tetrapyrrolic system, chlorins have one reduced pyrrole, and bacteriochlorins have two reduced pyrroles on the opposite side of the macrocycle [17,18]. The chromophores presented in Fig. 2 are characterized by strong absorption of UV-A radiation, weak absorption in the greenorange region, and most notably, they possess an intense absorp-  tion band in the red (typical for chlorins) or NIR (distinctive for bacteriochlorins) range of electromagnetic radiation [18]. While the results of research on the naturally-occurring macrocycles are undoubtedly valuable and inspiring, the easy access to a large amount of synthetic metallobacteriochlorins described in this review offers a number of new opportunities for the investigation of fundamental processes, fine-tuning of NIR optical properties, and the search for a range of photochemistry-based applications.
To gain access to stable and efficient bacteriochlorins, several de novo synthetic procedures have been recently developed [13,19]. Many of them afford bacteriochlorin with electron-withdrawing substituents. For example, geminal dimethyl groups lead to the stabilization of bacteriochlorin chromophore [20,21]. The other modifications that protect bacteriochlorins from oxidation include (i) insertion of appropriate metal ions into the macrocycle, (ii) presence of exocyclic rings in the macrocycle, and (iii) introduction of halogen atoms in the meso-tetraphenylbacteriochlorins [22]. The core structures of synthetic bacteriochlorins are shown in Fig. 3. Appropriate structural modifications enabled the synthesis of compounds containing very diverse substituents in different positions of the macrocyclic ring, which made it possible to obtain bacteriopyropheophorbides, bacteriopurpurinimides and tetraphenylbacteriochlorin derivatives, Fig. 3. There are also many N-fused porphyrins [23] or derivatives annulated with conjugated rings (e.g., anthracene [24]) as well as phthalocyanines [25,26] that absorb in the far red or/and NIR spectral region [24,27]. Especially, the strong NIR absorption of (metallo)bacteriochlorins is crucial for light-harvesting and possible biomedical applications [28]. Many authors reported the use of (metallo)bacteriochlorins that are capable of mimicking bacteriochlorophylls, either as separate molecules or as components of natural protein complexes [20,29]. The other applications of bacteriochlorins include: (i) solar energy conversion [30,31]; (ii) chemical or biological sensing [32][33][34]; (iii) (photo)catalysis [35]; (iv) spectroscopic analysis and design of fluorescent probes [36]; (v) molecular imaging [37][38][39], (vi) optical communication [40,41] and last but not least (vii) photodiagnosis (PD) and (viii) photodynamic therapy (PDT) [42,43], Fig. 4.

Basic photophysical processes, photochemical mechanisms and reactive oxygen species (ROS) generation pathways
Redox reactions involving the electron transfer between donor and acceptor molecules are the key processes determining the effective course of photosynthesis. The reduction of molecular oxygen by the photosystem I (PSI) in the Mehler reaction is a coupled process that results in the creation of ATP without NADPH [44]. It is therefore not unusual that ROS produced during photosynthesis are redox and functioning signals as well as necessary regulators of energy and metabolic fluxes [44][45][46]. In general, a photoactive molecule (photosensitizer, PS), such as bacteriochlorophyll a, possesses a stable electronic configuration in the singlet ground state (S 0 ) with electrons having opposite spins located in the highest occupied molecular orbital (HOMO). Following photon absorption, an electron may be excited to the lowest unoccupied molecular orbital (LUMO), resulting in the formation of the PS singlet excited states ( 1 PS*), Fig. 5.
Possible deactivation pathways of the PS excited states are illustrated in the modified Jabłoń ski diagram (Fig. 5). Higher PS vibrational states decay in a non-radiative way by releasing of heat. These fast processes (s~10 À11 -10 À14 s), include both vibrational relaxation (VR) and internal conversion (IC). If the PS molecule appears at the lowest vibrational level of the excited singlet electronic state, a spontaneous emission may occur. It leads to the allowed transition to the singlet ground electronic state -S 0 . The radiative process is called fluorescence and takes place in a nanosecond time regime. According to the Stokes rule, the maximum of the fluorescence (S 1 -S 0 transition) band is shifted towards longer wavelengths with respect to the maximum of the absorption (S 0 -S 1 transition) band. The transition of a PS molecule in the singlet excited state ( 1 PS*) to a triplet excited state ( 3 PS*) is not allowed due to the spin selection rule. However, it can be violated by spin-orbit coupling. If it happens, the spin reorientation occurs, and the molecule is transferred to the metastable triplet excited state (T 1 ) as a result of the process called intersystem crossing (ISC) [47]. Spin-orbit coupling usually occurs due to the so-called heavy atom effect caused by the presence of various metal ions, halogen and even oxygen atoms in the PS structure [48]. Consequently, these structural changes affect photophysical parameters such as the quantum yields of the intersystem crossing (U ISC ), fluorescence (U F ) and triplet state (U T ) as well as the triplet state lifetime (s T ). 3 PS* may either undergo phosphorescence (Ph) or, more preferably, can react with substrate molecule (e.g., molecular oxygen) in accordance with two main mechanisms of oxygendependent photosensitization reactions. Type I photochemical reaction includes an electron or a hydrogen transfer from a photosensitizer in either singlet or triplet excited state to some substrate molecules, wherease type II photochemical reaction involves direct energy transfer from 3 PS* to molecular oxygen in its triplet ground state ( 3 O 2 ) resulting in the generation of highly reactive singlet oxygen ( 1 O 2 , 1 D g ). Apart from these two basic mechanisms, there are also some references to type III or even type IV mechanisms. Those photosensitized reactions are independent of the molecular oxygen concentration. Most generally, they involve activated PS molecules and components of biological structures whose oxidation leads to the formation of reactive organic radicals [18,[49][50][51]. However, the most controversial issue still remains mechanism I. The process in which an electron transfer from the excited photosensitizer to dioxygen leads to the generation of superoxide radical ion (O 2 ÅÀ ). However, some authors suggest that the oxidized   form of PS should also be considered as type II reaction [52]. These mechanisms have been described in detail by us [18,48,51,[53][54][55][56] and by other authors [57][58][59], so they will be referred here only very briefly. Typically, a superoxide ion radical (O 2 ÅÀ ) is formed by a one-electron reduction of molecular oxygen. In the discussed photochemical processes, its formation is the result of the photoinduced electron transfer most likely from 3 PS*. Superoxide ion is quite a weak oxidizing agent in water. Nevertheless, in the subsequent step of oxygen reduction, hydrogen peroxide (H 2 O 2 ) is formed, which in turn in the presence of transition metal ions, in particular Fe 2+ and Cu 2+ , undergoes a Fenton reaction resulted in hydroxyl radicals production [60]. Hydroxyl radical ( Å OH), due to its high standard reduction potential (2.31 V), is one of the most powerful oxidants capable of reacting unselectively with the adjacent chemical compounds, including organic pollutants, biomolecules (proteins, DNA, lipids), and with almost any constituent of cells [56]. Although the photosensitization under biological conditions usually leads to a combined effect of both: type I and type II processes, there are some factors more or less suitable for each of them. Most frequently, the contribution of each individual mechanism is affected by the type of PS, its concentration, type of solvent/ environment in which the reaction takes place, and the amount of oxygen present in this environment. Low oxygen levels generally favor type I reactions, while type II usually occurs at higher oxygen concentrations [61].
The photophysical properties of naturally-occurring (bacterio) chlorophylls make them well adapted to support the energy transfer in photosynthesis. Unlike many endogenous pigments such as carotenoids, they do not undergo thermal relaxation [45], and thus, they can be maintained in the singlet excited state long enough for electron transfer as well as charge separation. However, these properties may create problems when light absorption exceeds photosynthetic light utilization. When excitation energy is not completely consumed during photosynthetic reactions, the lifetime of excited singlet state of bacteriochlorophyll ( 1 BChl*) increases in the light-harvesting antennae. This also enhances the probability of the ISC, resulting in the creation of a bacteriochlorophyll molecule in the triplet excited state ( 3 BChl*). It may cause a significant increase in the generation of described above reactive oxygen species: 1 O 2 , O 2 ÅÀ , H 2 O 2 , and Å OH. Therefore, the dissipation of thermal energy in bacteriochlorophyll systems can be understood as a protective mechanism responsible for preventing the formation of ROS instead of neutralizing them [45]. Nevertheless, it should also be noted that oxygen species generation is often associated with many beneficial effects. Moderate amounts of ROS are present in living organisms to guarantee normal metabolic functions and are involved in a number of signaling pathways in cells. A slight increase in ROS concentration enhances cell proliferation and increases the probability of cell survival. However, an excessive increase in their concentration can overcome the antioxidant capacity of cells and eventually lead to oxidative stress [62]. In normal cells, due to the presence of antioxidant enzymes, a small amount of exogenous ROS does not lead to any toxic effects. On the other hand, cancer cells, due to their dysfunctional metabolic activity, maintain ROS levels at the toxicity threshold, thus being in a permanent state of oxidative stress. An increase in ROS concentration in cancer cells triggered by drugs, radiation, photosensitizers, etc. may easily exceed their critical level, resulting in cell death. Thus, the fact that cancer cells are more sensitive to exogenous ROS has been used in numerous therapeutic strategies [63,64], including photodynamic therapy (PDT) and photodynamic inactivation of microorganisms (PDI) [53,65]. These two photochemistry-based modalities will be discussed in more detail later in this manuscript.

Physicochemical and spectroscopic properties of bacteriochlorophylls
The chlorophylls involved in plant photosynthesis and their synthetic analogs are the most studied pigments in terms of spectroscopic and photophysical properties [9,66,67]. Certainly less studied, probably due to their low stability, however essential for understanding fundamental processes and possible applications are bacteriochlorophylls (BChls) present in some photosynthetic bacteria [9,[68][69][70][71][72]. Compared to chlorophylls (dihydroporphyrins), bacteriochlorophylls are tetrahydroporphyrins with two reduced pyrroles placed on opposite sides of the macrocycle. The structures of bacteriochlorophylls a-e are shown in Fig. 6 (bacteriochlorophylls f-g are misnamed) [9,73].
Both chlorophylls and bacteriochlorophylls contain Mg 2+ as the central metal ion. It is known that the metal insertion into the tetrapyrrole ring changes its electronic structure and photophysical properties [75]. The progress in the reduction of pyrrole rings leads to intense absorption in the NIR. Whereas chlorophyll a absorbs at 662 nm and chlorophyll b at 644 nm, bacteriochlorophylls a, b, and g are characterized by strong absorption at 772 nm, 794 nm, and 762 nm, respectively [21]. Unfortunately, there are two major limitations concerning bacteriochlorophylls: (i) instability, including susceptibility toward adventitious dehydrogenation with the possible creation of the corresponding chlorin [72,76,77] and (ii) rigid and complete substitution of the macrocycle that reduces the possible semisynthetic transformations [21,42,78].
In order to overcome these limitations, a number of modifications are possible, including the peripheral substitutions of the macrocycle affecting their spectroscopic and photophysical properties. This modification results in a bathochromic shift of the low energy absorption band (Q band) towards NIR, while maintaining an appropriate energy level of excited states and lifetimes of these states long enough to allow sufficient photochemical reactions [79,80]. In that regard, synthetic chromophores have been also used for electron-transfer and energy-transfer processes. Thus, semisynthetic or synthetic bacteriochlorins can be an attractive alternative to bacteriochlorophylls. However, naturally-occurring bacteriochlorophylls still provide (Fig. 7) an excellent starting material for further modification [81].
Compounds obtained from bacteriochlorophylls were coordinated with several metal ions such as Zn 2+ , Cu 2+ , Pd 2+ , Co 2+ , Ni 2+ , Mn 2+ , Cd 2+ , and Pt 2+ [21,42,82,83] according to three main synthetic pathways [28]. The first one involves the coordination of a free macrocycle with Cd 2+ , followed by transmetalation. The other modifications consist of the introduction of Mg 2+ using a Grignard reagent and direct reaction with salts of appropriate metals. The insertion of metal ions into the macrocycles results in the significant change of their electronic structure and optical properties. These structural changes are particularly well described for porphyrins, for which an increase of symmetry from D 2h to D 4h is expected. This change leads to the degeneration of LUMO and a decrease in the number of Q bands in the electronic absorption spectra from four to two bands. In addition, one can observe a distinct batho-or hypsochromic shift of the Soret band, depending on the type of metal ion. Noteworthy, for Q bands, the introduction of metal ions always leads to their displacement towards shorter wavelengths [84][85][86]. For chlorins, however, no analogical changes in the electronic absorption spectra are observed. The coordination of metal ions does not influence the symmetry, but may cause a hypsochromic shift in the low energy absorption band assigned to the S 0 -S 1 electronic transition. For example, chlorophyll a (Mg 2+ chelated) possesses an absorption band at k max = 662 nm, whereas pheophytin a (free-base) absorbs at 667 nm in acetone [87]. Similarly, in the case of bacteriochlorins, the introduction of metal ions does not change the symmetry. Nevertheless, it leads to a bathochromic shift of the absorption band located in the red part of the spectra. Therefore, bacteriochlorophyll a (Bchl a) absorbs light at k max = 772 nm, whereas bacteriopheophytin a (BPhe a) at k max = 749 nm in diethyl ether [87]. More importantly, metal insertion significantly increases the molar absorption coefficient determined for long-wavelength Q bands [71].
A series of bacteriochlorophyll derivatives were also studied to determine the influence of the metal ions on the energy of electronic transitions and their redox properties [71]. Table 1 summarizes the spectroscopic (absorption and emission) and  electrochemical properties of bacteriochlorophyll a and its metal complexes. As expected, the electronic absorption spectra of all studied compounds reveal four characteristic absorption bands (B y , B x , Q x , and Q y ). All of the examined metal ions, but to a different extent, affect the energy of electron transitions. The most noticeable differences exsist in the energy corresponding to the Q x and Q y bands, while smaller differences are observed in the B x and B y bands. The most significant differences in the energy of S 0 -S 1 transitions were indicated for the BPhe complexes with Cu 2+ , Mg 2+ , and Mn 2+ metal ions. For these metal complexes, a large bathochromic shift (approximately 30 nm) of the Q y bands is observed [71,72,88,89].
The presence of metal ions in the macrocycle is crucial for the proper functioning of many natural processes. In general, the differences in the type of metal ion in the bacteriopheophorbide may significantly change the photophysical and photochemical properties. Magnesium derivatives (five or six coordination number) usually indicates the fluorescence quantum yield equal to 0.1, quite long singlet excited state lifetime (s s~1 0 ns), and relatively good quantum yield of intersystem crossing determined in diethyl ether (U ISC~0 .60 for Chl a and U ISC~0 .76 for BChl a) [84,89]. Zinc complexes (four or five-coordinate) are characterized by much lower fluorescence quantum yield (U F~0 .03), a shorter singlet excited state lifetime (s s~2 ns), and a higher ISC quantum yield [84]. These differences can be attributed to the heavy-atom effect [91], which is even more pronounced for complexes with heavier metals. For instance, the Pt-chlorophyll a triplet state lifetime determined in deoxygenated organic solvents is ten times shorter than for free-base and Mg 2+ and Zn 2+ coordinated chlorophylls [90]. Palladium derivatives (four coordinate) are also characterized by negligible fluorescence, the highest possible ISC quantum yield (U ISC~1 .00), and short triplet excited state lifetimes [86].
Analyzing the influence of different metals on the photophysical properties of bacteriochlorophylls, it can be assumed that the choice of Mg 2+ by nature and not Zn 2+ (which more easily forms complexes with tetrapyrroles), was not only motivated by the greater bioavailability of magnesium, but also to protect the photosynthetic apparatus against the excessive generation of ROS. This effect is quite different and more sophisticated for paramagnetic metal complexes (e.g., Fe 2+ , Cu 2+ , Ni 2+ and Co 2+ ) with tetrapyrrolic ligands than for Mg-derivatives. For such compounds, fluorescence is not observed at all, and highly predominant internal conversion can be detected. For instance, copper complexes with coordination number 4 are characterized by very short singlet excited state lifetimes, do not possess fluorescent properties, and, unlike other derivatives, have other properties highly dependent on temperature [85]. Also, the choice of paramagnetic metal ions by nature is considered as a protective mechanism against photochemical reactions that can eventually lead to adverse effects. Nevertheless, despite the lack of photochemical activity, this kind of metallopor-phyrins takes part in significant redox reactions and therefore participate in various biochemical processes [88].

Bacteriochlorophylls in the bacterial reaction center
The bacterial reaction center (BRC) with densely packed pigment molecules (e.g., chlorins and bacteriochlorins) surrounding the protein matrix from photosynthetic purple bacteria is well characterized in terms of their structure and spectroscopic properties [92,93]. The conversion through a network of wellarrangement pigments into protein complexes is an ideal model system for investigation of the structure-and function-relationship due to the efficient energy transfer and charge separation. Two-dimensional electronic spectroscopy (2DES) has emerged as a powerful method appropriate for examination the electronic coupling between chromophores [93]. The measurement of well design sequences of three laser pulses with determined emission frequency (t 1 , t 2 , t 3 ) allows the capturing of dynamics with high time resolution inside the system. This high resolution, nonlinear optical spectroscopic method allows studying the chemical and biochemical processes as well as interactions and dynamics of complex molecular systems. For instance, the observation of the light-harvesting antenna of photosynthetic microorganisms shows strong pigment-pigment interactions by capturing energy flow on a 2D frequency map, Fig. 8 [94].
Zigmantas et al. have studied light-harvesting complex LH3 from photosynthetic purple bacteria with B800-B820 bacteriochlorins, which absorb in 800 nm and 820 nm, respectively [92]. By using the 2D electronic spectroscopy, it was possible to measure the different population times that reveal ultrafast molecular dynamics within the B820 ring. Moreover, the asymmetry of the 2D spectrum of highly symmetric structure LH3 with the couplings between BChls eliminates cross-peaks by forbidding selected electronic transitions. LH3 contains nine identical subunits with 27 bacteriochlorophyll molecules in the symmetric ring-like light-harvesting structure. Closely packed 18 BChls of B820 ring are strongly coupled, while nine rings of B800 are relatively large separated subunits with exhibit weak intramolecular coupling [92]. 2DES can be suitable for the molecules for which the high-resolution crystal structure cannot be obtained. For instance, Ginsberg et al. have studied the light-harvesting complex 4 (LH4) expressed by photosynthetic purple bacteria in the lack of light with 2DES [95].
Nonlinear polarization-dependent spectra combined with theoretical modeling confirm the LH4 structure that consist of 8 identical protein subunits in an annular structure with a total of 32 molecules [96]. The significant efficiency of photosynthetic charge separation involving bacteriochlorophylls has inspired many scientists to recognize these processes at the molecular level. Konar et al. have also examined the model of the bacterial reaction center isolated from R. capsulatus (mutant with lack of ubiquinone). In these studies, the strong interaction between pigments located Table 1 Spectroscopic and electrochemical properties of natural and transmetalated bacteriochlorophyll derivatives [71,88,90].  [94]. The energy of couplings has been reported from the range~400-750 cm À1 for the special pair, to several cm À1 between H and P and HA and HB. Moreover, the excitation delocalization among the BRC is in accordance with the energy transfer and charge separation mechanisms. These data resolve Q y /Q x cross-peaks of most weaklycoupled BRC transitions with the two-color 2D spectroscopy [94]. The understanding of these naturally-occurring photosynthetic systems with efficient energy conversion influences the development of artificial light-harvesting devices.

Theoretical studies of bacteriochlorophylls properties
Parallel to experimental work, theoretical studies on the optical properties of bacteriochlorophylls containing divalent metal ions such as Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Ru 2+ , Rh 2+ , Pd 2+ , and Pt 2+ were conducted. Particular emphasis was given to the derivatives containing metal ions, which have already been used in anticancer therapy (for instance, Ru 2+ and Pt 2+ ). The main objective of these studies was to combine the cytotoxic effect of these metals with the spectroscopic properties of bacteriochlorophylls, enabling their application as PDT-photosensitizers. The theoretical calculations carried out using density functional theory (DFT) indicate that the geometries of the selected bacteriochlorophylls reveal many similarities [97]. The macrocycle is generally flat, and the introduced metal ions are placed in the plane of the ring, indicating the capability to insertion of metal ions with different sizes. Metal ions are most strongly bound to nitrogen atoms, which are derived from pyrrolic moiety. Based on the spectroscopic data present in the literature, the optical properties of the examined systems were also confirmed, Table 2 [97].
It was also concluded that all investigated complexes except Mg-BChl and Zn-BChl are stable in aqueous solvents, which enables their further investigation in biological systems. The mes-sage to be learned fom this study was that Ru-, Rh-, Pt-and Pdbacteriochlorophylls can be interesting photosensitizing agents for future therapeutic purposes [97]. These predictions turned out to be quite accurate because such types of compounds have been successfully tested in vitro, in vivo and even clinically for many years as potent photosensitizers for photodynamic therapy (PDT), which will be discussed later.

Bacteriopyropheophorbides and bacteriopurpurinimides
According to the procedure mentioned earlier, it is possible to convert the bacteriochlorophyll a extracted from Rb. sphaeroides into bacteriopurpurin-18. Bacteriopurpurin-18 in the next reaction steps is transformed into two types of bacteriochlorin derivatives: bacteriopurpurinimide and bacteriopurpurin p6, which are respectively equipped or not with a fused imide ring [98,99]. These types of long-wavelengths (metallo)macrocycles were developed by R.K. Pandey's group. For instance, the synthesis and characterization of a series of novel bacteriochlorins from 13 2 -oxobacteriopyropheophorbide as a starting material were reported.   Fig. 9 illustrates the general procedure for the synthesis of bacteriopheophorbide derivatives from bacteriochlorophyll a [100]. The absorption spectra of these bacteriochlorins showed exceptionally large shifts of the Q y absorption band towards nearinfrared (k max from 816 nm to 850 nm). Moreover, the determined molar absorption coefficients for these low-energy bands also possessed remarkably high values, even up to 136 800 M À1 cm À1 . In the following studies, the biological activity of these compounds and some of the analogs was extensively investigated and their excellent PDT efficacy against various types of cancer cells and tumor models was proven [42,101].
In the same research group, a library of new compounds called bacterioverdins was developed (Fig. 10). These molecules contain a fused six-membered methoxy-substituted cyclohexenone (verdin) as an isomeric mixture [100]. The most significant properties of bacterioverdins include absorption at long wavelengths (865-890 nm) and the favorable redox potentials that ensure both stability and ROS generation ability, not necessarily through energy transfer reaction. The comprehensive electrochemical studies were conducted, which part of the results in the form of cyclic voltammograms registered for each compound is demonstrated in Fig. 11. The most interesting results of these studies include the observation that three of the studied compounds, those with fused cyclohexanone ring systems, show a remarkably smaller HOMO-LUMO energy gap, and thus more readily undergo oxidation/reduction, compared to the previously investigated by authors bacteriochlorins [102,103]. As already mentioned, the greatest advantage of these compounds is their extraordinary absorption in the NIR range, reaching wavelengths of nearly 900 nm. Hypothetically, such compounds could be used in PDT to treat large, subcutaneous tumor tissue, because photons in this range are able to penetrate very deeply through tissues. Nevertheless, it should be noted that absorption at such long wavelengths exceeds the range specified for a phototherapeutic window (630-850 nm). This is naturally related to the energy limitation in the generation of singlet oxygen by such photosensitizers. However, appropriate redox properties of this group of compounds may facilitate photoinduced electron transfer reactions leading to the formation of oxygencentered radicals. Moreover, it seems that such compounds offer a completely different application potential, e.g., for the construction of optical materials and information storage.
R.K. Pandey et al. have also reported the impact of the insertion of metal ions, e.g., Zn 2+ , Cu 2+ , Sn 2+ , In 3+ on the photodynamic activity of examined complexes [104]. To investigate the effect of these metals on the photochemical properties, they prepared the corresponding complexes based on the pyropheophorbide framework (Fig. 12).

Pd-Bacteriopheophorbides
Compared to the corresponding free-based macrocycles, the Pt 2+ and Pd 2+ complexes are chemically more active and undergo more efficient ISC to achieve higher values of the triplet excited state quantum yields. Thus, combined with short excited-state lifetimes, they mainly act as type I photosensitizers generating oxygen-centered radicals in the biological environment [105]. The electronic absorption spectra of metallobacteriochlorins are similar to free-base analogues but slightly shifted either to shorter or more preferably longer wavelengths [106]. Most of the Pdbacteriochlorins photochemical properties make them valuable for the application in PDT.
Palladium containing BChl a derivatives such as padoporfin (Tookad, WST09) and padeliporfin (Tookad Soluble, WST11) presented in Fig. 13 are particularly interesting photosensitizers when compared to their free-base analogues [107][108][109][110]. First of all, pal- Fig. 9. Synthetic pathway for obtaining bacteriopheophorbide derivatives from bacteriochlorophyll a as a starting material. ladium bacteriochlorins, due to the heavy atom effect, are characterized by nearly unity quantum yield of ISC. Besides, as has already been demonstrated, bacteriochlorins coordinated with Pd 2+ have the most batochromically shifted long-wavelength absorption band compared to other metal complexes. Pd 2+ insertion into BChls results in the U ISC approaching 1 and an increased phosphorescence quantum yield under anaerobic conditions. The high value of U ISC is in line with the short lifetime of the singlet excited state. This brings the quantum yield of 1 O 2 equal to unity (U D = 1) in organic solvents [111]. Nevertheless, the relatively polar conditions (e.g., micelle/water solution) enhance the probability of charge-transfer processes between Pd-BChl a triplet state and molecular oxygen [111]. Besides, the large NIR absorption, desired redox properties and high value of U ISC determined for Pd-BChl derivative lead to a substantial formation of oxygencentered radicals upon a short irradiation time [111]. Studies on the influence of human albumin (HSA) on the generation of ROS by WST11 have shown that the photogenerated ROS are O 2 ÅÀ and OH Å radicals rather than singlet oxygen. This suggests that WST11 photocatalyzed the electron transfer from PS associated with HSA to the colliding oxygen molecules through several cycles [105]. The obtained results may suggest that the resulting WST11-HSA complex can be considered as a single reaction center, which photocatalysis redox reactions required for the formation of ROS, which are the appropriate cytotoxic agents in PDT [105]. The following research focused on palladium derivatives of BChl a led to the development of photosensitizers possessing remarkable photodynamic efficiency.

Zn-derivatives of bacteriochlorophylls
Semisynthetic and synthetic zinc complexes of bacteriochlorophyll a can be easily prepared by the replacement of the central Mg 2+ with Zn 2+ . Among the bacteriochlorophyll derivatives containing metals other than Mg 2+ , only Zn-chlorophylls have chemical features quite comparable to Mg-chlorophylls [112,113]. Nevertheless, Zn-derivatives are more stable than Mg-derivatives  and have been widely applied in artificial photosynthesis studies [114,115]. Zinc-bacteriopheophytins (Zn-(B)Pheo) are known to be good structural and functional models for naturally occurring bacteriochlorophylls [116,117]. It was found that Zn-BPhe may be used instead of BChl a in a special photosynthetic bacteria, e.g., an Addiphilium rubrum [118]. Moreover, Zn-containing bacteriochlorophyll has been introduced artificially into the isolated antenna proteins as the replacement of light-harvesting Mg-BChl or accessory Mg-BChl in the reaction center complex [70,113].
Yang and coworkers have synthesized and characterized the series of zinc-bacteriochlorins with the different number of electron-withdrawing groups as peripheral substituents [29]. The authors reported that their optical properties are not that different from free-base bacteriochlorins and naturally-occurring bacteriochlorophyll derivatives [118,119]. However, Zn 2+ insertion into macrocycle gives a lower fluorescence quantum yield (U F~0 .03), a shorter singlet excited state (s~2 ns), and a higher quantum yields of ISC and triplet state when compared with metal-free compounds [29]. Despite the lack of significant changes in the electronic absorption spectra, it should be recognized that for the series of Zn-bacteriochlorin with the different number of electron-withdrawing substituents, the differences in the values of molar absorption coefficients are significant [21]. Moreover, the zinc complexes are quite fluorescent, with quantum yields ranging from 0.08 to 0.20 with an average value of U F = 0.13, which is comparable to free-base analogs (U F~0 .15). The fluorescence lifetimes are also similar to free-base analogs (3.3-4.4 ns) [29]. The quantum yield of intersystem crossing determined for the Zn-bacteriochlorins (U ISC = 0.7) is somewhat higher than the average value determined for free-base analogs (U ISC = 0.5). This increase is again attributed to the heavy atom effect [29,120]. In conclusion, the synthetic zinc bacteriochlorins are characterized by improved photostability and the photophysical properties desired not only for medical applications but also for solarenergy conversion, light-harvesting and catalysis [28,[36][37][38]40]. Furthermore, they may also be applied in the development of conjugates and arrays designed with photosynthetic-like multipigment architectures [35,117,[121][122][123][124].

Ni-derivatives of bacteriochlorophylls
By changing the centrally coordinated metal ion in the bacteriochlorophyll macrocycle, it is possible to modify various properties not only in its ground but also in the excited states. It has been shown that derivatives containing Ni 2+ or Cu 2+ ions undergo rapid, nonirradiated conversion to the ground state without any photochemical activity. The Stolzenberg group delineated the structural and reductive chemistry of Pd 2+ and Ni 2+ octa-substituted bacteriochlorin derivatives for biologically significant differences in analog metalloporphyrins [125][126][127]. The synthesis of the Nicomplexes (Fig. 14) was accomplished by isobacteriochlorin metalations.
In the following studies, Lahiri and Stolzenberg have described the possible mechanisms by which Ni + -octaethylisobacteriochlorin anion can react with electrophiles [125]. They studied these processes in the context of the relevance of Ni + form in cofactor F430, which is a Ni-containing prosthetic group of methyl coenzyme M reductase. This biologically-relevant enzyme catalyzes the reductive cleavage of the thioether-cofactor in the final step of methanogenesis. It also catalyzes the reductive dehalogenation of chlorinated hydrocarbons by methanoic bacteria [125]. Nisubstituted bacteriochlorophylls may also be applied in the context of the light-harvesting antenna. For instance, Fiedor and coworkers have reported that the replacement of the central Mg 2+ in chlorophylls by Ni 2+ results in an ultrafast (up to femtoseconds) internal conversion, while maintaining the fundamental optical properties in the ground state [128]. Another potential application of these Ni 2+ complexes could be their use in the future as contrast agents for photoacoustic tomography.

Synthetic (metallo)bacteriochlorins
Despite the unquestionable success of palladiumbacteriopheoforbide, two main limitations of bacteriochlorophylls are related to (i) their relatively low stability and photostability, and (ii) the complete substitution pattern of the macrocycles, that reduces the possibility of semisynthetic transformations [29]. Synthetic bacteriochlorins offer an attractive alternative to semisyn-thetic derivatives of naturally occurring bacteriochlorophylls. The well-known synthetic methods of obtaining (metallo)bacteriochlorins include the hydrogenation and/or addition reactions (e.g., vicinal dihydroxylation) of suitable porphyrin or chlorin [20,29]. Nevertheless, these methods, depending on the existing substitutes, could lead to the formation of regioisomers and leaves the bacteriochlorin more prone to aerobic dehydrogenation [29].
The synthesis of the family of b-alkyl-substituted stable (metallo)bacteriochlorins was recently described. This procedure implies the modification of the standard transmetalation procedure using a strong base to protonate centrally placed -NH groups and then adding the appropriate metal salt. The set of synthetic bacteriochlorins bearing the different number of carbonyl groups were testes under the following conditions: Zn 2+ insertion to porphyrin ring and its further reaction in THF with a strong base upon the use of proper metal salt. In such a way several metal complexes can be obtained, namely: (i) derivatives with two aryl groups, M = Cu 2+ , Zn 2+ , Pd 2+ , and InCl (but not Mg 2+ , Al 3+ , Ni 2+ , Sn 2+ , or Au + ); (ii) bacteriochlorins with two carboethoxy groups, M = Ni 2+ , Cu 2+ , Zn 2+ , Pd 2+ , Cd 2+ , InCl, and Sn 2+ ; and (iii) bacteriochlorins with four carboethoxy groups coordinated with Mg 2+ [21,29]. Compared to free-base counterparts, these complexes are characterized by a 20-30 nm bathochromic shift of the low energy absorption Q band. The theoretical calculations also indicated a decrease in the HOMO-LUMO gap upon metalation [29]. It was shown that metalation induced a slight distortion from the planarity and a shift in the electron density on the frontier orbitals, particularly in the LUMO level [29]. The fluorescence quantum yields (U F~0 .01-0.0 3) and singlet state lifetime (s s~2 70 ps) of the metal complexes were diminished by order of magnitude when compared to corresponding free-base compounds. They also displayed a triplet state quantum yield U T~0 .9 and a triplet state lifetime s T~3 0 ls [21,29]. The metal insertion also increases the heavy atom effect by increased spin-orbit coupling, especially for the Inbacteriochlorins presented in Fig. 15.
Another set of metallobacteriochlorins presented in Fig. 16 was synthesized and thoroughly characterized by electrochemical measurements and DFT calculations. Their most important spectroscopic, photophysical and redox properties, namely the determined k max , the singlet and triplet state lifetimes, quantum yields of fluorescence and intersystem crossing (U F and U ISC ) as well as HOMO and LUMO orbitals energies and oxidation potentials, are summarized in Table 3.
The impact of metal insertion into bacteriochlorin macrocycle was described in detail for dicyano Zn-and Pd-bacteriochlorins and compared with their free-base analogs ((NC) 2 BC and BC, respectively), Fig. 17. Each bacteriochlorin indicats a typical electronic absorption spectra. Both modifications (metallation and introduction of cyano groups) lead to the noticeable shifts of the low-energy absorption Q y bands. The absorption band registered for metal-free dicyano-bacteriochlorin (NC) 2 BC in toluene is hypsochromically shifted (35 nm) compared to the non-substituted derivative [129]. The insertion of Pd 2+ results in a slightly bathochromic shift to 751 nm, while the Zn 2+ insertion causes an even more significant shift up to 761 nm, Table 4.
Moreover, in comparison to BC, metallobacteriochlorins are significantly less aggregated in water and are characterized by enhanced photostability due to the presence of metal ion and dicyano-substituents. BC, (NC) 2 BC and Zn 2+ complex indicate comparable singlet excited state lifetimes (s S = 3.9-4.1 ns) and fluorescence quantum yields (U F = 0.14-0.15) determined in toluene and methanol. Nevertheless, the fluorescence lifetime and quantum yield of (NC) 2 BC-Pd are remarkably reduced (s S = 23 ps, U F = 0.0008). The parameters determined for these complexes are consistent with those determined for other Pdbacteriochlorophyll derivatives [111] and imidazolium metalloporphyrins [129,130]. The intersystem crossing quantum yield determined for (NC) 2 BC-Pd reached unity, in contrast to U ISC = 0.63 determined for (NC) 2 BC-Zn and 0.43 for (NC) 2 BC. The values of U ISC obtained for unsubstituted BC (U ISC = 0.62) clearly demonstrate the influence of the cyano groups on the photophysical properties [129]. Consequently, the triplet state lifetimes in deoxygenated solutions are in the range from 80 ms to 170 ms for BC, (NC) 2 BC,   Table 3 Photophysical, spectroscopic and electrochemical properties of metallobacteriochlorins with the different number of electron-withdrawing groups [20,21,29,129].

Metal
Number of with-drawing groups  and (NC) 2 BC-Zn, but only 7 ms for (NC) 2 BC-Pd. The triplet state properties influence the ROS generation ability by these photosensitizers. It is reported that the Pd-derivative produces a relatively high amount of Å OH. The other compounds generate a smaller amount of Å OH than 1 O 2 and are, therefore, less susceptible to type I photoreactions. It is also demonstrated that the insertion of various metal ions influences the redox properties of studied photosensitizers. The first oxidation potential (E ox ) and higher reduction potential (E red ) of the four bacteriochlorins [BC, (NC) 2 BC, (NC) 2 BC-Zn, (NC) 2 BC-Pd] are summarized in Table 4. In general, the introduction of the two cyano groups into bacteriochlorin core results in the fact that the obtained derivative ((NC) 2 BC) is easier to reduce and more difficult to oxidize than the molecule without these substitutes (BC). Compared to free-base (NC) 2 BC, the palladium derivative is slightly easier to oxidize and also slightly more difficult to reduce. In contrast, the Zn 2+ complex is significantly easier to oxidize and, respectively, more difficult to reduce [129]. The importance of the redox potentials as factors influencing the balance between type I and type II photochemical mechanisms has been well established for dicyanibacteriochlorins and their porphyrin analogs [51,131]. As mentioned above, Pd 2+ complexes of these tetrapyrroles are harder to oxidize and easier to reduce (in a ground singlet and triplet states) than their Zn 2+ counterparts [129]. The redox properties of bacteriochlorins in the excited states affect not only their physicochemical (e.g., photostability) but also play the crucial role in the generation of reactive oxygen species via either energy or electron transfer reactions. The photoactivity of photosensitizers mediated by type I photoreactions may be partially derived from the transient reduction of the macrocycle. Thus, it is expected to be higher when photosensitizer undergoes a reduction in its excited state. Nevertheless, photobleaching of photosensitizers [37,132] is mainly related to photooxidation (instead of photoreduction) and photoaggregation (due to the p-cation radical formation). Thus, the photosensitizer, which is the hardest to oxidize, is also the most stable compound [133]. The improved photostability of photoactive compounds may be realized by introducing electron-withdrawing groups or coordination the central metal ion (e.g., Pd 2+ ) with high electronegativity into macrocycle [20,21,49,134]. This, in general, make photosensitizer harder to oxidize and easier to reduce. These statements may be supported by the redox potentials as well as energies of molecular orbitals determined theoretically in DFT calculation. Moreover, the molecular substitution pattern enables to control and tune the HOMO and LUMO energies, and redox potentials [20,135,136]. These relations were reported for the library of many synthetic photosensitizers, including chlorins, bacteriochlorins and their derivatives i.e. bacteriooxophorbines [38,111,137].
Recently, increased number of research reveals that type I photoreactions with the formation of hydroxyl radicals, superoxide and other oxygen-centered radical species play an important role in the photoactivity of photosensitizers. For instance, WST09 and WST11 are able to produce mainly superoxide anion and hydroxyl radicals [105]. The contribution of the type I mechanism was also determined for synthetic sulfonic and sulfonamide bacteriochlo-rins. It has been suggested that these bacteriochlorins may be expected to generate ROS via both mechanisms, whereas porphyrins undergo mainly energy transfer with the formation of singlet oxygen and they are in particular harder to oxidize than their reduced derivatives [129].
The pallet of metallobacteriochlorins described in this part ( Fig. 18) [21,29] will be discussed in the following sections in the context of their applications such as photodynamic therapy and photodynamic inactivation. The analysis of Fig. 18 suggests that from the synthetic point of view: (i) the unsubstituted bacteriochlorins without any electron-withdrawing groups indicate a limited possibility to obtain the metal complexes through reaction with a strong base; (ii) similar strong-base conditions can be applied for a broader scope of metals using bacteriochlorin bearing two carboethoxy substituents; (iii) for bacteriochlorins bearing carboethoxy or imide 2-4 substituents, the Zn 2+ can be inserted according to the standard porphyrin metalation procedure; (iv) for the derivatives bearing two carboethoxy substituents, reaction in the presence of strong bases allowed faster metalation process than the method involving the treatment with metal salt; (v) bacteriochlorin with 4 carboethoxy substituents gives the unstable Mg chelate under the strong base conditions; (vi) ortho-aryl substituents are known to significantly slow down the metalation of meso-tetraarylporphyrins [21].

Theoretical studies of synthetic bacteriochlorins with transition metal complexes
Parallel to experimental work, theoretical studies on the optical properties of bacteriochlorins and their metal complexes are also undertaken. L. Petit with collaborators performed theoretical calculations for a family of metallobacteriochlorins containing divalent transition metals including Mg 2+ , Zn 2+ , Mn 2+ , Fe 2+ , Co 2+ , Ni 2+ , and Cu 2+ . DFT and TDDFT were applied to study electronic properties and structural changes. Fig. 19 presents the energies of orbitals for each metallobacteriochlorin and the corresponding orbitals for free-base bacteriochlorin (FBBC). Interestingly, although the energy of the HOMO-LUMO gap was lower than the energy estimated for the isolated ligand, no dramatic changes are observed in comparison to FBBC. The presence of metal ions has little effect on the relaxation of FBBC degeneration according to the interaction with the ligand [138]. Solvent effects included in these calculations by the implementation of the polarizable continuum model (PCM) clearly indicate that both water molecules are coordinated with the central metal ion. Unlike to evolution ion radius, the determined values are a bit reduced in the transition from Mn 2+ to Ni 2+ , and then they come back to Zn 2+ . It may suggest that stronger adhesive interaction for metal than Fe 2+ to Ni 2+ concerning Mn 2+ , Cu 2+ , and Zn 2+ . These calculations also revealed that Zn 2+ is characterized by the lowest interaction with the ligand, while Ni 2+ shows the greatest effect. The studies show the strongest ion interactions of the central metal with nitrogen for Fe 2+ , Co 2+ , and Ni 2+ , whereas for atoms with half-filled (d 5 Mn) or filled (d 10 Zn) orbitals, weaker interactions are observed. Special attention was paid to the less energetic part of the spectra, Table 5. The result of these calculations clearly shows that the metalation causes red shift of Q x and blue shift Q y bands, while comparing to the free-base bacteriochlorin [139]. The intensity of electronic transitions indicates a slight increase in the molar absorption coefficients after the metalation of the free-base compound. All of these metal complexes except the Fe 2+ complex possess an energy Fig. 19. Theoretical calculation of orbital energy levels for (metallo)bacteriochlorin derivatives. Adapted from [138].  Table 5 Calculated energies of electronic transitions (eV) along with oscillator strengths in parenthesis for FBBC bacteriochlorin and its metal complexes [138].  [140,141]. The preparation of these compounds consists of the following steps: (i) the synthesis of appropriate porphyrins by modified nitrobenzene method that involves the condensation of pyrrole with the appropriate aromatic aldehyde in the presence of nitrobenzene in acidic conditions; (ii) chlorosulfonation of the obtained porphyrins; (iii) either hydrolysis or reaction with amines to obtain respectively: hydrophilic (sulfonated) or amphiphilic (sulfonamide) halogenated porphyrins; (iv) diimide reduction of these porphyrins to bacteriochlorins; the metalation with the desired metal ion (as an example, the introduction of Mn 2+ and Mn 3+ is presented in the Section 4.2.2) [22,[141][142][143]. The first attempt to the synthesis of the halogenated sulfonamide and sulfonated bacteriochlorins involved a modified Whitlock method, including the use of toluene as a solvent and cat-alyzed by a relevant organic base. In the following years, it was possible to develop an alternative synthetic procedure in accordance with the sustainable chemistry approach. This method is based on obtaining compounds under solid-solid conditions, without the use of any solvents. Its key advantages are: efficiency, simplicity, low environmental impact and the ability to provide a library of photostable NIR-absorbing dyes with various polarities and characterized by favorable physicochemical properties and enhanced biological performance [144]. The general scheme of the synthesis of halogenated bacteriochlorins and their metal complexes are presented in Fig. 20. Fig. 21 shows the structure of selected compounds belonging to this family, together with their electronic absorption spectra [145]. Modifications within the tetrapyrrolic ring lead to the significant changes observed in their electronic structure. The typical freebase porphyrin with D 2h symmetry is distinguished by the presence of the intense absorption band in the blue part escribed as the Soret band and additional four Q bands in the range of larger wavelengths (500-700 nm) corresponding to transitions between vibrational levels of much lower intensity. The change in the symmetry of the system caused by the metal ion insertion results in a change in symmetry of the system to D 4h , which results in the reduction of the number of Q bands in the absorption spectra from four to two. These Q bands are formed as a result of the difference in dipole vectors between the HOMO-LUMO levels. The Soret band is most often observed as one absorption band due to the slight energy differences in the case of porphyrins. If the energy differences between the respective levels are significant, then the degeneration of the LUMO orbitals occurs, resulting in the splitting of the Soret band (Fig. 21) [146,147]. The reduction of subsequent pyrrole rings in the macrocycle results in the destabilization of the HOMO level and the LUMO orbitals. With the increasing degree of reduction of the pyrrole rings, the energy of transition between the occupied orbital a 1u (p) and the unoccupied orbital e gy (p*) increases, whereas between the orbital a 1u (p) and the unoccupied orbital e gx (p*) does not change. The effect of these modifications is manifested in the decrease in the energy of the HOMO-LUMO transitions following: porphyrins > chlorin > bacteriochlorins. The decreased HOMO-LUMO energy gap is also correlated with the redox properties of porphyrins, chlorins, and bacteriochlorins.
Together with the degree of reduction, the HOMO level energy increases, so that the susceptibility to oxidation changes in the following series: porphyrin < chlorin < bacteriochlorin (lowering the oxidation potentials), without changing the reduction potential [148].
Despite the favorable optical properties of bacteriochlorins (k max = 740-780 nm), their significant limitation was low stability, susceptibility to oxidation, and, as a consequence, reduced possibilities of ROS photogeneration. Therefore, one of the objectives of the research discussed here was to apply desired modifications within the macrocycle, resulting in the pallet of photostable NIRabsorbing compounds, which through appropriate redox and photophysical properties shall generate a high amount of ROS according to both types of photochemical reactions (type I and type II mechanisms). The first possible modification consists of the introduction of phenyl rings into the tetrapyrrolic system in the meso positions, which act as steric hindrance, which ensures compounds with greater stability and more amphiphilic character. The introduction of halogen atoms (-F, -Cl) to the ortho position of the phenyl rings leads to spin-orbital coupling, increasing the yield of the intersystem crossing. Moreover, these electron-withdrawing groups stabilize the macrocycle against its oxidation. Achieving large values of the efficacy of the ISC process also creates the ability to control the mechanisms of photosensitization, and thus the efficacy and type of generated ROS. The parameter allowing to assess the effectiveness of the spin-forbidden transition is the spin-orbital coupling constant (f), which for the hydrogen atom is f = 0.24; in the case of chlorine: f = 586, and for fluorine f = 269 [149]. It appears that the presence of a chlorine atom allows achieving two times higher heavy atom effect. Nevertheless, the introduction of fluorine atoms into the structure of the tested compounds allowed to obtain optimal photophysical properties (improved photostability, generation of triplet states), while maintaining the fluorescence capabilities facilitating biological experiments being carried out (e.g., subcellular localization, studies of pharmacokinetics and biodistribution [22,25,142,[150][151][152]. The summary of properties determined for the library of halogenated bacteriochlorins is presented in Table 6.
In addition to the effect on the photophysical properties of the studied molecules, the introduction of fluorine atoms into the structure also plays an important role in pharmacological properties [153]. Fluorine exhibits high biocompatibility and is a widespread pharmacologically active compound [151]. The greater therapeutic efficacy of compounds containing fluorine atoms in the structure compared to non-fluorinated compounds has been described by other authors [150]. The next possible modification of the new synthetic bacteriochlorins involves their functionalization with sulfonic or sulfonamide groups as well as coordination with certain metal ions [154]. Substitution of sulfonic and sulfonamide groups allows to control the hydrophobicity of molecules and to increase their stability by the effect of steric protection [54,140,142,155]. The introduction of substituents with higher electron-receptor properties leads to increased hydrophobicity of photosensitizer, which affects its interaction with biological membranes [156,157]. Moreover, the presence of sulfonamides in the bacteriochlorin structure results in a remarkable increase in the selectivity towards cancer cells and tumors as well as photodynamic efficacy due to the higher lipophilicity of such photosensitizers [158].

Manganese complexes of halogenated bacteriochlorins
Manganese(III)-porphyrins are characterized by catalytic reactivity, the possibility of photoreduction to Mn 2+ and suitability for the use as photoacoustic (PA) references [159,160]. Moreover, they may be used in the production of effective piezophotonic materials [161]. The described in the literature extraordinary spec- troscopic and photochemical properties of Mn 3+ -porphyrins have encouraged the synthesis of novel Mn 3+ and Mn 2+ coplexes of halogenated sulfonamide bacteriochlorins. The subsequent steps of the synthesis of these manganese complexes of halogenated tetraphenylbacteriochlorin are illustrated in Fig. 22 [162].
The properties of the electronic states of the Mn-based compounds are significantly different from free-base bacteriochlorins or their complexes with diamagnetic metals, Fig. 23. It was reported that the Mn 3+ complexes reveal eight electronic transitions between various electronic states, including both: (i) p->p* transition, typical for bacteriochlorin and (ii) ligand-to-metal transitions. Among them, the transition corresponding to the strong absorption band at~830 nm deserves particular attention. The lowest-energy transitions involve direct excitation from the singlet ground to triplet excited state (e.g., S 0 (p,p*) ? T 1 (p,p*) transitions). Nevertheless, the interaction of metal (d, d) orbitals and the orbitals of the macrocycle (p,p*) is not as significant as observed for manganese(III) porphyrins [163,164]. This smaller mixing of orbitals diminishes the intensity of the CT transitions and increases the triplet excited state lifetime. Interestingly, the bacteriochlorin abbreviated (Cl)MnCl 2 BMet 2 , as for a paramagnetic metalloporphyrin derivative, is characterized by the relatively long lifetime of the triplet excited state (s T~5 70 ps) and a quite substantial quantum yield of the triplet state (U T~0 .2). Hence, the applicabil-  ity of (Cl)MnCl 2 BMet 2 in photocatalysis or photovoltaics may be considered. However, the efficient internal conversion in this manganese(III) bacteriochlorin and its quite intense absorption at k max = 829 nm (e~50,000 M À1 Ácm À1 ), where tissues are transparent indicate that this class of compounds can be suitable for biomedical applications such as photothermal therapy or even photoacoustic tomography [165].

Morpholino-and lactone-based bacteriochlorins
The conversion of pyrroline to morpholine moieties in nearplanar meso-tetraarylbacteriochlorins results in ruffled monoand bis-morpholinobacteriochlorins (Fig. 24) characterized by broadened and bathochromically shifted electronic absorption spectra. Observed changes in the optical properties have been assigned to a combination of two independent factors: (i) dihedral angle between morphine and pyrrole moiety and (ii) molecular planarity [139]. Conformational changes characterize bacteriochlorin derivatives containing one or two non-pyrrolic heterocycles with oxygen atom into the framework. Nonplanar conformations induced by peripheral substituents in porphyrin may cause redshift of Q band even up to 50 nm [166]. The pyrrole reduction also distorted the planar macrocycle, especially in two-reduced bacteriochlorin, which increases ring flexibility. Insertion of oxygen in the reduced domain with sp 3 hybridization leads to even more remarkable conformation changes than in free-base bacteriochlorin. Nevertheless, the general trend between distortion and spectrum is red-shifted [167]. However, in some cases of structural changes, morpholino-bacteriochlorins indicate blue-shifted Q band caused by space-through interaction of electron-donating oxygen with aryl groups.
Mazzone and co-authors have also provide computational studies on the morpholino-bacteriochlorins, which entails a distortion of the entire macrocycle derived from the insertion of the oxygen atom between the two pyrrolidone carbons [168]. Computational studies of the morpholino-derivatives with the introduction of a different number of the phenyl group in the meso position also show a notable effect on the conformation. Compared to dihydroxy-dimethoxy bacteriochlorin, a bathochromic shift of 50 nm is observed for the morpholino-bacteriochlorin, which is in line with spectroscopic experiments performed in solutions. The crucial requirement for ideal photosensitizer is the energy gap between the ground singlet and triplet excited state higher than 0.98 eV adequate to the energy gap between triplet and singlet oxygen. Considering the optical properties of these compounds, including the HOMO-LUMO gap (Q band), as well as the energy gap between the singlet-triplet excited state, the expanded  bacteriochlorins bearing morpholino groups can be proposed as photosensitizers act via type II photochemical mechanism with the singlet oxygen formation [168].
Nevertheless, the extension of the p-conjugation system may cause undesirable changes in solubility (via p-p stacking), stability and aggregation. Thus, the modulation of aromaticity in the tetrapyrrolic ring is a significant challenge, especially in reduced porphyrinoids. Another group of extended bacteriochlorin-like compounds with a long-wavelength absorption in the NIR is represented by porpholactones. The introduction of an oxygen atom into the macrocycle ring in b-peripheral position may extend p-system without causing disadvantages mentioned above. Additionally, the isomerization of b-oxazolone may tune the electronic ground states, singlet and triplet excited states as well as the redox potentials. The porpholactone derivatives are characterized by strong absorption in the NIR, shifted similarly to naturally-occurring hydroporphyrins. The introduction of the oxazolone group in the macrocycle leads to, besides the p-system extension, lower symmetry (C s ) and stabilization of the HOMO and LUMO levels. The isomerization may also significantly influence the photophysical and photochemical properties of porpholactones. In this case, similarly to naturally-occurring (bacterio)chlorophylls, cis/trans isomers indicated the changes in the absorption spectra. Moreover, it provides high efficiency in energy storage in photochemical processes, e.g., mediated by cis-isomer of chlorophyll d derived from cyanobacteria, which possess lower molar absorption coefficient than chlorophyll a (non-isomeric). Moreover, compared to the Q y absorption of chlorophyll d (cis-isomer), its trans-isomeric form (chlorophyll f) indicates the stronger red-shift in the NIR (ca. 18 nm), Fig. 25 [169].
Zhang and coauthors have developed the novel synthesis method of the porpholactone using desired ruthenium salt and oxone. The oxidation of tetrapentafluorophenylporphyrin results in porpholactone (F 20 TPPL, 85%) and porphodilactone (F 20 TPPDL, 15 %) derivatives with both trans-and cis-isomers [169]. It was also demonstrated that regioisomeric bacteriodilactone-based derivatives with non-bonding electrons are more aromatic than their typical bacteriochlorin analogs [96,171], with red-shifted of Q y band (19 nm) and increased absorption comparable with chlorophyll f and chlorophyll d [172]. Although porpholactones indicate the similar optical properties to chlorophylls, their coordination capabilities are comparable with porphyrin derivatives. In this respect, porpholactones and their metal complexes (Zn 2+ , Pd 2+ , Pt 2+ , Fig. 26) [172] play an important role as a molecule for many applications including photodynamic therapy, optical imaging, and well as a novel sensing materials (see chapter 6). Moreover, they might be applied in catalysis, photocatalysis and even electrocatalysis.
The regioisomerisation of metal complexes of porphodilactones may also cause changes not only on absorption spectra but may also alter the triplet excited states. For instance, trans-isomers of Pd 2+ and Pt 2+ complexes showed red-shifted and longer-lived phosphorescence than cis-analogs (Table 7) [170]. The discussed porpholactones and their metal complexes might be applied as small molecule activation catalysts and photo-or electrocatalysts. Computational study of the ground and excited states for lactonebased bacteriochlorin demonstrated the change of their stability as well as has an impact on the absorption spectra [96]. The stabilities and strong electronic influence of series regioisomeric porphodilactone (Fig. 26) due to the charge delocalization, was also reported.

Basic principles of PDT
Photodynamic Therapy (PDT) is a photochemistry-based medical strategy, dedicated mainly for cancer treatment, but recently also widely used in the inactivation of localized bacterial infections Fig. 25. The influence of the orientation of b-substituents on NIR absorption of naturally-occurring chlorophylls (Chl b, d, f) and porpholactones (cis-and trans-F 20 TPPDL). Adapted and modified from [170]. [53,174]. Three nontoxic components are needed to initiate photochemical reactions: a photosensitizer (PS), a visible/near-infrared light and molecular oxygen commonly present in the cells or tissues. These components together result in the generation of highly cytotoxic reactive oxygen species. Fig. 27 shows the subsequent steps of the PDT procedure. The administered photosensitizer is activated, and then undergoes several photophysical processes and photochemical reactions. As discussed in Chapter 2, PS in the triplet excited state is able to transfer an electron/hydrogen atom or energy to another molecule, e.g., O 2 to generate ROS. These species are involved in the destruction of biological structures, leading to oxidative stress within the tumor according to three related anti-cancer mechanisms: (i) direct cytotoxic effect resulting in apoptosis, necrosis and/or autophagy (cell self-healing process), (ii) the indirect closure of tumor blood vessels, and (iii) stimulation of the immune system to local and systemic responses, and induction of pro-inflammatory processes eventually leading to the development of antitumor immunity [49].
The contribution of each mechanism depends on the applied light dose, drug dose, drug-to-light interval (DLI), oxygen concentration in the tumor tissue and applied irradiation margin [18]. The immune response against cancer cells is probably the most important aspect of PDT because it creates the possibility of not only destroying the primary tumor but also preventing the occurrence of metastases and finally taking a big part in the complete cure of the disease [175,176]. On the other hand, the critical step in the PDT studies is to develop a photosensitizer (PS) characterized by desired chemical and biological properties [132,177]. For instance, it should indicate a strong absorption in the phototherapeutic window, high values of U ISC and U T , as well as long triplet state lifetime [18].
Photogenerated ROS causes damage to biological structures at the site of the PS localization, which consequently leads to tumor destruction. An important issue is the determination of whether and how the metal insertion or other substituents introduced into the macrocyclic ring affect the ability to control photoinduced electron and energy transfer reactions [55,154,157,178,179]. These studies are focused on the idea of how modifications influence the ROS generation in cancer cells [18,55,153,157,179,180], and improve their selectivity towards diseased areas [178,[181][182][183][184]. The most frequently studied PDT-photosensitizers are tetrapyrrolic macrocycles, including bacteriochlorins and metallobacteriochlorins [58]. Clinically used photosensitizers, depending on their polarity, may either bind to the plasma proteins (hydrophilic compounds) or accumulate in different cellular compartments (compounds with higher lipophilicity) [58,174,185]. In most cases, the cell nucleus is not the primary target for photosensitizers, whereby PDT does not induce genotoxic effects compared to classical chemotherapy and radiotherapy [186,187].  The ability of PS to accumulate in cells and their intracellular localization not only determines the appropriate choice of formulation and incubation time but also allows to indicate the primary site of photodamage [188]. For many years, singlet oxygen has been considered the most important toxic agent responsible for photoinduced cellular damage in PDT. Despite a relatively short half-life, it is characterized by quite a large interaction range (its activity reaches the radius of 20-550 nm) [189,190]. For comparison, hydroxyl radicals generated according to type I mechanism possess a radius of only 1 nm, but are more reactive than singlet oxygen, which in turns is more selective oxidant. In the development of new photosensitizers, besides the determination of physicochemical and photochemical properties, pharmacokinetic studies in vivo are essential to evaluate absorption, distribution, metabolism, and excretion of potential drug candidates. The selectivity of photosensitizers can be determined using tumor-to-skin (T/S) and tumor-to-muscle (T/M) ratios. However, many features can affect the tissue distribution and pharmacokinetic profile after intravenous PS administration. The most critical of these are PS charge, formulation and parameters of applied therapeutic protocol including drug dose, tumor model, drug-to-light interval (DLI), administration route, light dose and type of the light source. Thus, the direct comparison of the results obtained from different laboratories is difficult and should be untertaken with coution [191]. The pharmacokinetics and biodistribution largely determine the choice of DLI [158,180,184]. The current state of knowledge allows to distinguish three basic variants of protocols. The first one is characterized by short DLI from 5 to 15 minutes and is named vascular-targeted photodynamic therapy (V-PDT). The second one (cellular-targeted photodynamic therapy, C-PDT) involves the selective delivery of PS to cancer cells. For this protocol DLI reaches >12 h, but usually involves 24 h, 72 h, or even 96 h from PS administration to tumor irradiation. The last protocol is focused on the targeting endothelial cells (endothelial cells-targeted photodynamic therapy, E-PDT). It is characterized by an intermediate DLI (usually 3 h) when the photosensitizer is found mainly in the blood and endothelial cells with simultaneous accumulation in tumor cells [153,180,192].

Bacteriochlorophyll derivatives as naturally-occurring and semisynthetic photosensitizers
Bacteriochlorophylls are characterized by stronger absorption in the NIR than chlorophylls, which makes them theoretically more suitable candidates for anticancer PDT [193]. However, naturallyoccurring Chls and BChls are not relevant for photomedicine because they are not enough photostable. Moreover, they contain a hydrophobic chain in the structure (phytyl), which hinders their safe application and further localization in cancer cells. Nevertheless, through the use of structural modifications within the macrocyclic ring discussed in the Chapter 3, it is possible to increase their application potential. 5.2.1.1. Bacteriochlorophylls as PDT sensitizers. Many bacteriochlorophyll derivatives were tested both in in vitro and in vivo conditions. The photoactivity of BChlide a and BChl-Ser was evaluated in M2R mouse melanoma cells [194]. It was also reported that BChl derivatives could be active even under hypoxic conditions, acting mainly via type I photoreactions with hydroxyl radicals formation [195]. Moreover, replacement of the Mg 2+ with Pd 2+ in BChl structure significantly increased the photodynamic effect in various cells, decreasing the LD50 value by order of magnitude without any change in the dark toxicity [196]. BChl-Ser was applied as PS against M2R melanoma and showed quite short elimination time from the body (16 h), which prevents skin photosensitivity [194]. The optimized PDT protocol involved drug dose of 20 mg/kg BW, light dose of 108 J/cm 2 , and tumor irradiation immediately after PS administration. [197,198]. The described protocol, focused mainly on the tumor vasculature disruption and blood flow stasis, was characterized by significant efficacy, reaching over 80% cures of mice [197]. The PDT activity was confirmed in other tumor models, for instance in SD sarcoma-bearing rats [199,200]. BChl-Sermediated photodynamic efficacy was also tested in combination with hyperthermia [199,200]. The tumor response to PDT with BChl-Ser as photosensitizer was further controlled with a specific intertissued oxygen microsensor. The rapid reduction of tumor oxygen concentration was observed due to the efficient ROS production during irradiation [201].
The first BChl-based compound, which was entered into the clinical trials, was Pd-bacteriopheophorbide (padoporfin, Tookad, WST09, see Fig. 13 and chapter 3.2.2) [202]. Vascular-targeted PDT (V-PDT) with Tookad led to a significant therapeutic efficacy against human prostate cancer as well as bone metastases. In the discussed PDT protocol, NIR laser irradiation (k = 770 nm) with a light dose of 90 J/cm 2 was carried out immediately after intravenous PS administration at a dose of 4 mg/kg BW [109]. The therapeutic efficacy of Tookad was also verified towards prostate cancer in dogs with the therapeutic regimen including NIR light irradiation (k max = 763 nm), light dose of 200 J/cm 2 and a quite high drug dose of 5 mg/kg BW [81]. Tookad-V-PDT was also effective in the treatment of HT29 colon carcinoma (multidrug-resistant and wild types) as well as C6 glioma tumors growing in mice [203][204][205][206]. WST09 (padoporfin) has proven to be a promising photosensitizer first in preclinical [207], then in early clinical trials for PDT of prostate cancer [208,209] and was finally applied in clinics in 2004.
The data obtained from patients recruited to phase I suggested that the skin photosensitivity related to V-PDT is negligible under clinical protocol involving PS dose of 0.1-2 mg/kg and irradiation (k = 763 nm) at ligh doses from 10 to 360 J/cm 2 [210]. The drug and light doses escalation studies were also performed in patients with locally recurrent prostate tumors receiving external beam radiotherapy (EBRT) and the tumor response were monitored with Gd-enhanced MRI [208]. The phase I/II clinical trial conducted by Gertner et al. indicated the safety as well as overall efficacy and pharmacokinetic data of padoporfin-V-PDT [211]. The parallel trials conducted by Arumainayagam et al. also involved the drug and light dose escalating studies. The patients (Gleason score 7 and PSA <20 lg/L) were subjected to V-PDT with 2 mg/kg padoporfin administered in a 20 min i.v. infusion and irradiation with 763 nm diode laser (100-300 J/cm 2 ). The optimal light dose able to efficient tumor destruction was indicated as 200 J/cm 2 . Moreover, the short-term side effects that appeared on the urinary tract were insignificant. Nevertheless, in one case, the cardiovascular event and stroke were noted as an adverse result of hypotension. It was supposed that these consequences are related to the padoporfin formulation containing Cremophor EL (CrEL), and thus the clinical trial has been canceled [212].
Nevertheless, due to the high efficacy in PDT protocols directed against tumor vasculature and positive results of phase I and II clinical trials, padoporfin was replaced by the derivative with higher hydrophilicity (padeliporfin, Tookad Soluble, WST11, see Fig. 13) [105,110]. WST11 has a hydrophilic character that originated from the presence of negatively-charged -SO 3 H groups [78,108,213]. The introduced modification ensures that this derivative does not require the use of any special delivery vehicles for its intravenous administration. Tookad Soluble is capable of forming the non-covalent complex with human serum albumin (HSA) after intravenous administration. Thus, after injection, it can exist in two forms (free-unbound and bound with HSA) and remain in the circulation system, which can reduce the photodynamic treatment directed to the tumor vasculature [105,214]. Nevertheless, WST11 indicated the highest concentration in blood immediately after i.v injection. During the time (up to 1 h) this concentration increase in other tissue and finally the PS is eliminated from the body [110]. Moreover, WST11 does not induce systemic toxicity and its short life-time guarantees the rapid removal from the body that also reduces the possibility of phototoxic side effects. In contrast to WST09, WST11 does not accumulate in endothelial cells, and its photodynamic action is directed only toward blood vessels [110]. The structural changes between WST09 and WST11, as well as in their formulations, result in differences in their pharmacokinetic profiles and tissue distribution. WST09 indicated the high activity in V-PDT due to maximal concentration in plasma achieved after ca. 5 min and was eliminated from the body already after 3 h. The effective concentration of WST09 in other organs, i.e., in the liver and kidneys was initially high and then decreased over time. These trends in tissue distribution, as well as body clearance, suggest the undesirable binding to the RES system, which is characteristic for hydrophobic drugs [215]. In consequence of short circulation time in the body, PDT with Tookad Soluble requires very short DLI. It should be highlighted, that both photosensitizers (WST09, WST11) indicated no capacity to accumulation in tumor and surrounding healthy tissue (skin or muscle). This effect provides their high selectivity towards tumor vasculature and tumor microenvironment (extracellular milieu) [214,215].
The phase I/II clinical trials with water-soluble padeliporfin also revealed its safety and tolerability in patients with prostate cancer [216]. Azzouzi and coworkers have performed a comprehensive analysis of data obtained in phase I/II (NCT00946881) and phase II trials (NCT00707356, NCT00975429) from the patients (Gleason score 7, PSA <10 lg/L) after V-PDT (4 mg/kg, 10 min i.v. infusion, 753 nm, 200 J/cm 2 ) [217]. The primary outcome was long-term negative biopsies determined in 68.4% of examined population (>100) and 80.6% of patients received the hemiablation with light density index (LDI, the parameter characterized the ratio of the length of applied fiber optics in cm to the volume of treated area expressed in mL) higher or equal to 1. Based on these results, it was also indicated that in both groups, the prostate necrosis was relatively high and PSA concentration decreased significantly [203]. In phase II studies, the hemiablation in patients was also observed during V-PDT with transperineally delivered radiation [218]. In other studies, the necrotic area in ca. 90% of the treated prostate tumors was detected using MRI. Moreover, V-PDT with padeliporfin was presented as a highly tolerable and safe procedure and the related adverse effects like prostatitis and other urinary tract problems (e.g., haematuria) were reported only in single cases [217]. During phase II investigation, the optimization of V-PDT protocol was performed. For this reason, the men with PSA <10 lg/L were subjected to the treatment and the most prominent therapeutic outcome within the negative biopsy was observed in 83 % of patients. Moreover, from the statistical point of view, the average scores of the International Prostate Symptom Score (IPSS), as well as the quality of patients' life, indicated a significant improvement with no further changes in the International Index of Erectile Function score (IIEF-5). Moreover, it was demonstrated that the patients treated with lower dose of padeliporfin (2 mg/kg) revealed no significant therapeutic effect, while patients subjected to a higher dose of PS (6 mg/kg) were characterized by necrotic areas in surrounding organs [219]. Moreover, the histological changes and long-term consequences examined 6 months after V-PDT revealed relatively low adverse effects [220].
In the next step, Steba Biotech [221], as well as Azzouzi et al. [223] have undertaken a multicenter phase II clinical trial to select the optimal PDT conditions and estimate the safety and healthrelated quality of life [222]. During II and III phases of clinical trials, the standardized procedure of V-PDT mediated with padeliporfin against prostate cancer was drawn up [204,217,223]. The efficacy and feasibility of salvage radical prostatectomy (RP) after padeliporfin-V-PDT was assessed based on the data from: NCT00707356, NCT00975429 and NCT01310894, respectively. For further evaluation of V-PDT against prostate cancer, the interventional III phase of trials has been developed in the group of patients recruited from Mexico, Panama and Peru. In the follow-up investigation, the multicentre, randomized controlled and open-label phase III (ca. 400 patients) was realized to compare the results of padeliporfin-V-PDT with active surveillance in the prostate cancer treatment [221,224,225]. The main characteristics of the described clinical trials are summarized in Table 8.
Padeliporfin received EMA approval in 2017, is marketed in Mexico and is under consideration by FDA. Finally, in November 2017, it was successfully approved for clinical practice in Mexico, Israel, as well in the European Union (in 31 countries) and the European Economic Area (EEA) [230]. Furthermore, besides earlystage localized prostate cancer, padeliporfin is also being developed against other various types of tumors. In 2018 phase I clinical trials in urogenital cancer (recurrent, second-line therapy or greater, inoperable/unresectable) in the USA were started (NCT03617003) and were launched for prostate cancer in the United Kingdom. Moreover, at the beginning of 2020, Steba Biotech informed about planning a phase III clinical trial in prostate cancer in USA (NCT04225299) [229].
Recently, V-PDT with WST11 is applied with the combination of neoadjuvant or other treatment modalities. For instance, the possibility of potentiation of antitumor response by V-PDT with PD-1/ PD-L1 immune checkpoint inhibition against melanoma was described [230]. V-PDT was also combined with androgen deprivation therapy (ADT), which is the most effective strategy for metastases [231]. Moreover, it may be a promising strategy because transcriptome profiling provided an upregulated androgen response pathway following WST11-V-PDT [231]. Thus, it can be hypothesized that targeting this pathway by a combination of ADT with V-PDT should be more effective compared to PDT or ADT monotherapies [231]. It appeared that the heptyl ether substituents with relatively high lipophilic characters are the most potent PDT photosensitizers resulting in 80% cures of treated mice. All of the described data lead to the conclusion that photosensitizers with longer alkyl ether carbon chain indicate the highest PDT efficacy [98].
In other studies, 13 1 -(4-aminobutylcarbamoyl)bacteriochlorin methyl ether derivatives with different-length of aliphatic chains were synthesized and characterized. These photosensitizers were photodynamically active even at nanomolar concentrations against various types of epithelial cells. Moreover, the in vivo studies performed on S37 sarcoma tumor-bearing mice confirmed their potential as PDT agents. The most significant therapeutic effect was achieved after the application of 2.5 and 5 mg/kg BW drug doses. After the application of the optimized PDT regimen, the Table 8 The summary of clinical trials with padoporfin and padeliporfin in V-PDT against prostate cancer. Adapted and modified from [226]. 100% of tumor growth inhibition, as well as long-term cures of animals followed up to ca. 4-months post-PDT was observed [213].

Synthetic bacteriochlorins as PDT photosensitizers
In contrast to naturally-occurring bacteriochlorophyll derivatives, many PDT photosensitizers used in clinical practice are hydrophilic or amphiphilic. A characteristic feature of hydrophilic photosensitizers (>2 negative charges) is the lack of the ability to penetrate biological membranes. They can accumulate in the cells mainly via endocytosis or by photochemical internalization (PCI) [232,233]. In many studies, it has also been shown that hydrophilic photosensitizers (e.g., padeliporfin) [234,235] can accumulate in lysosomes. Hydrophilic photosensitizers are easier to prepare in aqueous solutions for their in vivo administration [177]. After intravenous injection, they interact with plasma proteins, accumulate mainly in the tumor microenvironment and lead to significant vascular effects.
The significant photodynamic activity of selected bacteriochlorins may be related to their molecular design and substitution pattern. The lipophilic character of these photosensitizers provides their specific accumulation in critical organelles such as endoplasmatic reticulum (ER) and mitochondria. Moreover, their spectroscopic and photophysical properties (discussed in detail in chapter 4), e.g., high molar absorption coefficient (~120,000 M À1 Ácm À1 ), determined for the NIR absorption band and the high triplet state quantum yield make them very efficient against resistant cancers including pigmented melanoma [236,237].
In the following studies, selected synthetic metallobacteriochlorins including free-base unsubstituted bacteriochlorin (BC), its dicyano-derivative (NC) 2 BC, as well as corresponding zinc (NC) 2 BC-Zn and palladium (NC) 2 BC-Pd complexes were also investigated in vitro against HeLa cells (see Fig. 17) [212]. The photophysical and photochemical properties of these compounds were described in chapter 4. The effect of dicyano peripheral substituents, central metal ion (Zn 2+ , Pd 2+ ) and CrEL formulation on the photodynamic activity was examined in biological conditions. Their efficacy (DLI = 24 h, light dose = 10 J/cm 2 ) may be arranged in the following order: (NC) 2 BC-Pd > (NC) 2 BC > (NC) 2 BC-Zn % BC. Moreover, these bacteriochlorins indicate different subcellular accumulation in organelles such as mitochondria, lysosomes and ER. Noteworthily, the best efficacy of Pd 2+ complex is most likely related to damages of mitochondria and lysosomes caused mainly by photogenerated hydroxyl radicals. These results confirm that structural modification of bacteriochlorin-backbone via metal insertion and introduction of electron-withdrawing substituents (such as cyano moieties) provide a promising template for the development of more active NIR-absorbing photosensitizers [239].

Halogenated sulfonic and sulfonamide bacteriochlorins.
Probably the most versatile photosensitizers are amphiphilic compounds -containing both hydrophilic and hydrophobic substituents in their molecular structure [186,187]. They accumulate both in individual cell organelles, as well as membrane structures and intracellular space. Thus, they can be used either in protocols targeting peritumoral vessels (V-PDT) or directed to cancer cells (C-PDT) [57,240,241]. Bacteriochlorins possessing halogen atoms in the structure (chapter 3.2.1.) exhibit a low tendency to aggregation, strong absorption in the phototherapeutic window, enhanced stability and efficient formation of long-lived triplet states. Furthermore, sulfonic/sulfonamide substituents allow the control of the hydrophilicity/hydrophobicity balance and provide additional protection against oxidation [22,54,142]. Halogenated and sulfonamide bacteriochlorins have been successfully tested against various types of cancer cells [154,157,[180][181][182]184,242]. They have proven to be non-toxic in a wide concentration range (0.001-50 mM), which make them potentially suitable for biomedical applications. It has also been shown that studied photosensitizers efficiently accumulate in cancer cells depending on the molecular structure of PS, incubation time as well as cellular phenotype [54,178,243]. The time-dependent cellular uptake of chlorinated sulfonamide derivatives increased steadily over time and reached a maximum after 18-20 h of incubation [158]. Similar results were obtained for fluorinated sulfonamide bacteriochlorins, while the more polar, negatively-charged sulfonated bacteriochlorin (ClBOH) indicates a relatively low ability to accumulation in cancer cells [243]. The cellular uptake of the studied compounds seems to be facilitated by their well-balanced amphiphilic character. Sulfonamide bacteriochlorins are uptaken by cancer cells in amounts ten times higher than their sulfonated analogs [155]. Moreover, amphiphilic (sulfonamide) photosensitizers revealed a high degree of localization in critical cellular targets such as ER and mitochondria, whereas hydrophilic (sulfonated) compounds were found mostly in the lysosomes. The photodynamic effect mediated by these bacteriochlorins led to pronounced morphological changes in cells integrity. Most of them can be related to the hallmarks of apoptotic cell death, i.e., cells shrinkage and pyknosis were visible by light microscopy [186,187,244]. Furthermore, it was reported that fluorinated sulfonamide bacteriochlorin is able to induce autophagy [245] and immunogenic cell death in vitro [246,247].    29. It is clearly demonstrated that with regards to their substitution pattern -the most effective photosensitizer is difluorinated sulfonamide bacteriochlorin (F 2 BMet), in contrast to monochlorinated sulfonamide one, which exhibit the lowest efficiency (ClBEt). In addition, there is no correlation between observed photodynamic efficacy and, determined for each photosensitizer, singlet oxygen quantum yields, but it is closely related with their ability to generated oxygen-centered radicals.
The in vivo results obtained for the library of synthetic halogenated bacteriochlorins indicate that they are nontoxic even at high drug doses (up to 100 mg/kg BW) and preferentially accumulate in organs such as the spleen, liver and tumor. It was established that the maximum concentration of halogenated bacteriochlorins accumulated in murine melanoma tumors is observed after 24 h post-PS administration [180,181]. For the photodynamic therapy in vivo, the tumors with a diameter of approx. 0.5 cm were selected [154]. After 24 h post-PS or i.v. injection of photosensitizer, tumors were irradiated with laser light (650 or 750 nm) with a power ranged from 50 to 100 mW/cm 2 . It has been shown that excitation of the sulfonated bacteriochlorin with laser light from the near-infrared range results in significant inhibition of tumor growth after photodynamic therapy in all tested experimental groups of animals [154,155,178,180,242]. Moreover, after photodynamic treatment with analogs sulfonated chlorin, the complete curative effect can also be observed. Noteworthy is that chlorin derivative was found to be more effective in vivo, despite the much inferior photodynamic effect determined in vitro and less favorable spectroscopic properties [181,183]. This effect was explained by relatively higher redox potentials examined for this chlorin derivative, and in consequence, its higher photostability (the quantum yield of photobleaching for this compound was lower by one order of magnitude when compared with bacteriochlorin analog). The higher stability of chlorin derivative also influences its pharmacokinetics and interactions in biological environment after administration in vivo. This photosensitizer indicates to be more biocompatible and shows improved pharmacokinetic properties characterized by longer circulation time in the body and thus, higher ability of accumulation in the tumor tissue [158,181,184]. The photoactive drug designed for cellular tageting should lead to high T/S and T/M ratios and be characterized by high bioavailability. Both requirements are revealed for chlorin derivative; thus, it can be suggested that investigated chlorin derivative as photosensitizer allows achieving acceptable profiles of tolerabil-ity, safety and high therapeutic efficacy in photodynamic therapy of murine tumors [183]. Despite such therapeutic success, in further studies, even more effective photosensitizers were sulfonamide derivatives [155,158]. In all PDT-treated mice with a bacteriochlorin-based photosensitizer, complete tumor regression was observed with no normal tissue damage in the immediate vicinity of the tumor. The explanation of this therapeutic outcome may be related to higher photostability of sulfonamide derivatives and more favorable spectroscopic properties [54,142]. The lead compound from this library is difluorinated sulfonamide bacteriochlorin (F 2 BMet, redaporfin). The high phototherapeutic efficacy of redaporfin was confirmed in various tumor models (e.g., CT26, A549, LLC, B16F10) [54,155,156,180,184,192]. The doseescalation and toxicological studies demonstrated the low toxicity of redaporfin and its CrEL-based formulation after application with optimized PDT protocol (DLI = 15 min; drug dose = 1.5 mg/kg; light dose = 74 J/cm 2 ; laser power = 130 mW; margin Ø = 13 mm). For instance, redaporfin-PDT of a CT26 tumor-bearing BALB/c mice provided a significant long-term antitumor effect with a high cure rate of 83% [154,242]. The mechanisms of photodynamic activity of redaporfin were studied in detail and this photosensitizer reached clinical trials in the treatment of head and neck cancers (NCT02070432) [248,249]. Parallel to the clinical trials, there are ongoing studies on the use of redaporfin for PDT treatment of multidrug resistanct (MDR) tumors [250] and in the combined therapy with immune checkpoints inhibitors.

Vascular effects of redaporfin-based PDT
Based on the results of the comprehensive research described above, redaporfin has been chosen for the evaluation of vascular effect in PDT. Studies carried out on S91/I3 tumors growing in DBA/2 mice indicated that treated tumors are characterized by an extended hypoxic fraction (mean pO 2 level in tumors with a volume of 50-100 mm 3 reached value about 2 mm Hg). Cellulartargeted PDT with F 2 BMet (DLI = 72 h) initially leads to disturbances in the structure and function of vessels, and then to the normalization of vascularity (about nine days) in the PDT-treated area. In contrast, therapy directed against tumor vasculature (DLI = 15 min) destroyed the vascular structure in the tumor and led to complete inhibition of their functionality for about 48 h after irradiation. These effects consequently led to strong tumor hypoxia (pO 2 = 0 mm Hg) [251]. Results of following in vivo studies per- formed on LLC tumors also showed that V-PDT led to the development of long-term antitumor response as a consequence of vascular occlusion and tumor starvation. In the case of C57 mice bearing LLC tumors, the complete curative effect was observed in 67% of animals followed for over a year after PDT treatment. This significant effect can be mediated by, i.e., bacteriochlorin-based PS properties as well as real-time monitoring of oxygen distribution, its intertissued concentration and controlled density of tumor vasculature before and after treatment, Fig. 30 [192]. These data reveal that in the case of LLC tumors, V-PDT results in more significant therapeutic success than other PDT protocols with longer DLIs (e.g., E-PDT or C-PDT).

The antitumor immune response after bacteriochlorin-based PDT
It is recognized that PDT can trigger rapid inflammatory responses that are essential for the activation of antitumor immunity and long-lasting protection against metastasis. The generation of post-PDT antitumor response involves the participation of various mechanisms based on the secretion of cytokines and mediators associated with the inflammatory processes, increased level of neutrophils in the blood combined with neutrophil recruitment to the treated areas, and induction of acute-phase proteins and activation of the complement system [252][253][254][255]. PDT-induced inflammation is associated with the influx of the cells of the immune system to the treated area, and these cells contribute to the development of an immune response that recognizes cells that survived the therapy [256,257].
The effect of V-PDT on the immune system was investigated with Tookad Soluble as photosensitizer and the antitumor immune response was indicated as the main factor responsible for the observed overall therapeutic efficacy [258]. The long-term systemic antitumor immunity involves both cellular (e.g., activation of lymphocytes) and humoral pathways, including changes in the antigens recognized by the serum immunoglobulins (IgG) and B cells recruitment. The anticancer activity was defined as crossprotective against different tumors (CT26 and 4T1 tumor-bearing BALB/c mice), suggesting that V-PDT-mediated production of both type tumor-derived antigens may be related to endothelial origin. Thus, it can be suggested that locally applied V-PDT can be combined with other antitumor strategies (e.g., immunotherapy) for the enhancement of antitumor immunity in the treatment of both local tumors as well as distant metastases [258].
Pd-Bacteriopheophorbide-mediated PDT (5 mg/kg, 650-800 nm light dose of 360 J/cm 2 ) against established C6 tumors (rat glioma) growth in CD1 nude mice resulted in complete tumors regression. Moreover, PDT decreased the kinetics of tumor growth and reduced possible lung metastases which was not observed after surgery [107]. V-PDT against renal cell carcinoma also potentiated systemic antitumor immune response by PD-1/PD-L1 immune  [192]. checkpoint inhibition [230]. Mice bearing primary renal tumors were treated with either V-PDT alone, with PD-1/PD-L1 antagonistic antibodies alone, or with a combination of V-PDT and antibodies. The results confirmed that only V-PDT in combination with systemic PD-1/PD-L1 pathway inhibition, resulted in the regression of primary tumors, prevented growth of lung metastases, and significantly increased prolonged survival. It also demonstrates the crucial role of local immune modulation with PDT in combination with PD-1/PD-L1 pathway inhibition for the genera-tion of systemic antitumor response [230]. Elimination of primary tumors and control of metastasis after PDT was also reported for redaporfin. The antitumor immune memory of CT26 tumorbearing BALB/c mice cured with the V-PDT was compared with the group cured by surgery, Fig. 31A. It was observed that all the mice treated with surgery developed tumors in contrast to animals rechallenged with CT26 cells in the contralateral thigh. Accordingly, in this group, 67% of redaporfin-PDT treated mice remained cured for more than three months. This immune memory effect also reveals that the optimized PDT protocol was effective in stimulating the immune system via activation of T cell adaptive immunity and long-lasting protection against metastases, Fig. 31B [242].
More recently, it was reported that the PDT regimes with polarity-tunable bacteriochlorin lead to different biological responses with the activation of distinct immunological components, Fig. 32 [156]. Various degrees of local inflammation is observed after different PDT regimes. The analysis of selected cytokines present in the blood before and 24 h after the PDT showed increased IL-6 concentration in plasma after PDT, as expected from the observed inflammation. The enhanced concentration of KC and MIP-2 chemokines after PDT suggests that tumor response may involve an influx of neutrophils to the tumor. Moreover, the optimal protocols with sulfonamide bacteriochlorins (F 2 -BMet-V-PDT and Cl 2 BHep-C-PDT) are characterized by increased secretion of MIP-1, which is also involved in neutrophilic inflammation. The optimal protocols also indicated the increased level of TNFa activity and decreased VEGF expression, especially after anti-vascular treatment. These findings offer the molecular basis of how the changes in selected biomarkers after PDT contribute to mounting an adaptive immune response capable of inducing immunological memory and recognition of tumor cells that have survived the local PDT treatment [156].

Photodynamic efficacy enhancement by PS functionalization
The direction of the research consisting of the construction of phototherapeutic systems dedicated to targeted therapies has also contributed to the use of nanotechnology in the design of photosensitizers with higher pharmacological activity and simultaneous minimization of side effects (allergic reactions, long retention time in the body). Additionally, the use of new formulations and selective systems for the supply of bacteriochlorins allows to increase their (photo)stability, accumulation in cells, reduces the aggregation in the biological media, and eventually increases in vitro and in vivo efficacy [259][260][261][262]. They also affect significantly the pharmacokinetics and pharmacodynamic parameters ensuring pre-ferred tissue distribution in the body [263][264][265][266]. Moreover, they influence not only the photophysical properties allowing the control of the electron and energy transfer processes but also the localization of induced photodamages. The use of various functionalization pathways enables determination of the relationship between the structure and physicochemical properties as well as biological activity of bacteriochlorins in the context of their application in photomedicine [53,174,175].

Polymeric micelles and nanoparticles
Several drug delivery nanosystems have been designed to target many tetrapyrrolic-based photosensitizers towards cancer cells and tumor microenvironment [265,267,268]. Considering the hydrophobicity of bacteriochlorophyll derivatives, Gomes et al. have evaluated the photophysical and photochemical behavior of BChl a, when loaded into a drug delivery system (DDS) such as poly(D, L-lactide-co-glycolide) (PLGA) nanoparticles. Nanoparticles of biodegradable poly(D, L-lactide-co-glycolide) loaded with BChl a were successfully synthesized with high yield and satisfactory incorporation efficiency using a solvent evaporation technique. The authors described the method suitable for the encapsulation of BChl a and enabling safe in vivo administration [269]. BChl a has shown to be a promising alternative to the molecules currently being used in PDT due to its spectroscopic properties, such as a NIR absorption at 770 nm (in toluene) and 782 nm in an aqueous solution of nanoparticles, respectively. Moreover, the increased singlet oxygen generation quantum yield (U=0.31) is related to the ROS formation mainly via type II photoreactions.
The functionalization of redaporfin with Pluronic Ò copolymers increased the effectiveness of the therapeutic regimens [180]. The pharmacokinetic profile and photodynamic efficacy of redaporfin-CrEL, redaporfin-F127, and redaporfin-P123 in animal models (B16 melanoma-bearing C57BL/6J mice) were determined. The results of the conducted research indicate the possibility of a significant increase in the selectivity of the applied photosensitizer and the possibility of obtaining a 100% long-term cure for animals after the application of the F 2 BMet-P123-V-PDT therapeutic protocol [180].

Lipoproteins
Lipoproteins represent the class of biocompatible, nonimmunogenic complexes containing the lipids and proteins, which are ideal for efficient drug loading and delivery. Many lipoprotein-based nanoparticles create the opportunities for enhanced therapeutic and theranostic effect through mimicking the shape and structure of endogenous lipoproteins. Due to the small size (>30 nm) of the low-and high-density lipoproteins (LDL and HDL, respectively), they can avoid the reticuloendothelial system (RES) and penetrate the tumor tissue more deeply. In addition, they are characterized by longer circulation time in the body. Moreover, lipoproteins are useful in cancer treatment because many tumors (e.g., colon, breast, prostate) indicate increased expression of LDL receptors. Furthermore, LDLs may also be re-targeted to folate receptors that are more specific to the selected type of human tumors [270,271].
Marotta and coworkers have developed the LDL-based delivery system for lipophilic bacteriochlorophyll derivative -bacteriochlorin e6 bisoleate (abbreviated as r-Bchl-BOA-LDL) for its used in PDT. Bis-oleate or bis-stearate moieties have also been applied for loading of hydrophilic dyes (like tricarbocyanines) and Gdchelating substituents into the phospholipid monolayer of LDL [272]. The biological activity profile of r-Bchl-BOA-LDL was investigated in vivo on human hepatoblastoma G2 (HepG2) tumors. When compared with untreated control mice, the kinetics of tumor growth was significantly delayed after PDT with PS dose of 2 mmol/ kg and different light doses (125, 150 or 175 J/cm 2 ). Moreover, assessed phototoxicity against tumor as well as minimized healthy tissue damages confirm that r-Bchl-BOA-LDL may be used as an effective PDT agent [272].

Metal-based nanoparticles
Nanoparticles (NPs) and, more specifically, noble metal NPs are widely-known agents for many biomedical applications, including promising drug delivery systems and diagnostics tools. In general, noble metal NPs (e.g., Au, Ag) indicate the high surface-to-volume ratio and appropriate, tunable spectroscopic properties that may be modulated to desired wavelengths. Such tuning is possible due to their shapes (nanorods, nanoshells, etc.), size (up to 100 nm), composition (core/shell or alloys) and potential surface modification and functionalization [273]. Pantiushenko and coworkers have described the comprehensive characteristics of a new type of materials containing bacteriochlorophyll isolated from non-sulfur bacteria Rh. capsulatus. This derivative posseses the disulfide moiety derived from lipoic acid. Such types of bacteriopurpurinimides-based photosensitizers are characterized by aurophilicity (a phenomenon referring to the aggregation of gold complexes). Due to the possibility of the formation of S-Au bonds, such properties allow obtaining the conjugates with gold nanoparticles (NP-Au). The conjugation is related to the redshifted NIR absorption (at k max = 824 nm) and fluorescence emission at k max = 830 nm. The biological tests performed in vivo in rats bearing M1 sarcoma indicated the favorable biodistribution of PS-Au estimated by real-time dynamic fluorescence imaging. The authors have reported that PS-Au conjugates exhibited a longer circulation time in the blood than free bacteriochlorophyll a as well as enhanced accumulation in tumor tissue that makes it a more effective nanoformulated theranostic agent [274].
The other type of NPs are magnetic nanoparticles consist of a metal/metallic oxide core, encapsulated in an inorganic or a polymeric coating shell. This shell renders the properties of the particles, such as their biocompatibility and stability, as well as may act as a support for many biomolecules. Their magnetic properties enables their application as MRI contrast agents, hyperthermia agents, magnetic wand vectors controlled by magnetic field gradient towards a selected location similar to targeted drug delivery systems [275]. The bacteriochlorin-loaded magnetic nanoparticles dedicated to personalized MRI-guided PDT were recently described. The bacteriochlorin derivative with a positive charge and various-lengths linker between the amine group and macrocyclic ring was efficiently encapsulated in HSA coated metal NPs via electrostatic and hydrophobic binding. After successful encapsulation, magnetic NPs-HSA-PEG nanoparticles provide long-term stability as well as effectiveness in delivering PS to cancer cells in both in vitro and in vivo systems. Moreover, this kind of conjugates reveals that they can be applied in in vivo MRI imaging for efficient drug delivery into the tumor milieu [276].
The same research group performed a primary screening of bacteriochlorin-based photosensitizers containing aminoamide, propyl, and peripheral carbohydrate substituents. The authors performed the comprehensive spectroscopic characteristics of these compounds, determined their stability as well as dark and phototoxicities against HepG2 cancer cells [277]. All studied photosensitizers except the bacteriochlorin substituted with carbohydrate moiety in the exocycle E were not cytotoxic in the dark and led to the strong photodynamic effect. The highest photoactivity was noticed for aminoamide-bearing photosensitizer, for which two hours of incubation led to the IC50 in the range of 17-49 nM). Based on all of the determined physicochemical and biological properties for this group of agents, O-propyloxime-N-propoxybac teriopurinimide methyl ester was ranked as the most promising agent in the treatment of tumors [277]. These studies were further continued by Plotnikova et al. in order to develop the formulation for O-propyloxime-N-propoxy-bacteriopurpurinimide methyl ester. The CrEL micellar emulsion was indicated as a promising formulation of this photosensitizer and was of great interest for further studies in the field of PDT [277].

Metal-organic frameworks (MOFs)
Nanovehicles designed for therapeutics have gained increasing interest, as these systems allow to avoid many disadvantages of traditional treatments. A new class of hybrid organic-inorganic materials is represented by metal-organic frameworks (MOFs), containing selected metal ions and organic bridging ligands. MOFs may be promising nanoplatform for selective drug delivery, with the opportunity of high drug loading, biocompatibility as well as wide-spectrum functionality and applications [278]. The hydrophobic character of many tetrapyrrolic derivatives may be a reason not only for insufficient selectivity towards tumors but also may cause aggregation and reduce the efficacy of PDT. However, this makes them very attractive examples of organic ligands in the MOF structure [279]. The use of tetrapyrrole-based MOFs designed for PDT application have been investigated. However, in many cases, they are conjugated with other nanoparticles forming dual-agents for synergistic strategy (including PDT-photothermal therapy, PDT-radiotherapy). Recently, bacteriochlorin-based MOF was prepared as a NIR-absorbing dual-mode photoagent for photoacoustic imaging PAI-guided PDT [279]. This MOF contains the 5,15-di(p-benzoato)bacteriochlorin (H 2 DBBC) as blocks and central clusters, including Hf 6 (m 3 -O) 4 (m 3 -OH) 4 (abbreviated as DBBC-UiO). This hybrid material was dedicated to the treatment of hypoxic tumors due to the ability of ROS generation via both types of photoreaction with singlet oxygen as well as oxygen-centered radicals formation, Fig. 33. It was reported that the DBBC-UiO MOF could generate superoxide anion within a severe hypoxic microenvironment, and, that the part of the generated amount of O 2 ÅÀ is converted into hydroxyl radicals via SOD-induced catalytic process upon irradiation with NIR light [279]. This effect suggests that the photoactivity of DBBC-UiO may also be oxygen-independent and effective against hypoxic tumors. In the following studies, the ability of MOF application in tumor-specific PAI was verified and these data indicate that it could also be used as an agent for diagnosis of cancer with deeper tissue penetration and increased resolution [279].

Application of (metallo)bacteriochlorins in photodiagnosis (PD) and imaging
The growing interest in cancer detection focuses on the ability for better visualization of selected probes that target diseased tissues. Real-time imaging of designed fluorescent probes enables the improvement of both basic and preclinical research. The photons from the near-infrared region are able to penetrate biological tissues more efficiently than visible light [3]. However, at wavelengths k>950 nm, these effects are reduced due to increased absorption of water and lipids. Thus, the phototherapeutic window used in PDT (630-850 nm) is not optimal for photodiagnosis, considering the higher background signal from tissue autofluorescence as well as limited light penetration through tissue (1-2 cm). Therefore, in photodiagnosis the optical window covers the range from 700 nm to 900 nm [280]. In this region, the endogenous pigments do not absorb and, consequently, do not interfere with the NIR-fluorescent molecules [281][282][283][284]. Moreover, NIR photons indicate a lower ability to generate ROS due to the lower energies in this part of spectra [3,6]. The spectroscopic properties of bacteriochlorin derivatives indicated the strong absorption band between 700 and 800 nm is of particular interest because it allows higher tissue penetration depth and avoids the optical interferences with endogenous pigments. The ability to distinguish multiple-color molecular markers during fluorescence imaging is one of the most appropriate feature. Such an effect can be achieved by using an appropriate probe or in combination of selected labelling-agents. These fluorophores should indicate significant Stokes shifts derived from wellseparated absorption and fluorescence emission bands. Moreover, this property reduces the signal of the scattered excitation light, and consequently, to improve the quality and resolution of registered images [36,285].
Several NIR fluorophores, like cyanine dyes (Cy5, Cy7), indocyanine green (ICG), Alexa Fluor compounds as well as tetrapyrrolebased fluorochromes have been extensively developed for biomedical applications [38]. Despite the fact that some commercial probes are based on these structures, several new classes of labelling agents, including bacteriochlorophyll-or bacteriochlorin framework are still being developed. Besides the favorable characteristics, some of the studied bacteriochlorins demonstrated limitations for their successful application. The major limitations are due the reduced stability of bacteriochlorins. As previously described, many efforts have been taken to modulate the stability of BChls, including peripheral modification of the macrocycle, the introduction of the serine or halogen atoms as substituents, insertion of central metal ion, or encapsulation in selected drug delivery systems. For instance, Ptaszek and coworkers have described several promising dyads with potential use as optical imaging probes [37,38]. The chlorin-bacteriochlorin, as well as BODIPYbacteriochlorin complexes are able to take part in the efficient energy transfer, and thus, are considered as promising fluorescence probes. In such a chlorin-bacteriochlorin system, chlorin acts as an energy donor and bacteriochlorin plays an energy acceptor role. Excitation of the chlorin leads to stronger fluorescence emission derived from bacteriochlorin [37,38]. This complex exhibits several desirable properties, including strong absorption and fluorescence, narrow bands with FWHM <20 nm and large Stokes shift over 85 nm. Moreover, these chlorin-bacteriochlorin dyads are characterized by relatively long-lived excited singlet state ca.~5 ns [38]. Additionally, the fluorescence properties can be tuned by a structural modification and changes in the peripheral substitution patterns. In other studies, the BODIPY-bacteriochlorin complex was synthesized and characterized, Fig. 34 [286].
In this type of bioconjugates, the BODIPY molecule, contrary to chlorins, possesses increased hydrophilicity and high molar absorption coefficient in the green range of spectra. However, due to the bulky aromatic structures and overall hydrophobicity, Fig. 33. The scheme of the synthetic procedure for metal-organic framework DBBC-UiO formation and its potential application for photoacoustic imaging and photodynamic therapy. Adapted and modified from [279].
PEG formulation was used to improve the water solubility and alter their pharmacokinetics in vivo [287]. To achieve enhanced selectivity and active targeting of fluorescent probes, many efforts were undertaken, including the conjugation with antibodies. Unfortunately, probe-antibody complexes (e.g., panitumumabbacteriochlorin conjugates) were characterized by increased hydrophobicity and were prone to aggregation [287].
Moreover, the synthesis and characterization of the other bacteriochlorin derivatives (named as NMP4 and NMP5) were reported in terms of their use as targeted photodiagnostic agents [285]. These compounds, after excitation with a green light indicate the fluorescence emission in the NIR (k~739 nm for NMP4 and k 770 nm for NMP5, respectively) [285]. In the subsequent studies, these bacteriochlorins were conjugated to galactosyl-human serum albumin (hGSA) or glucosyl-human serum albumin (glu-HSA), which are the molecules with affinity to H-type lectins. Thus, these complexes were enabled to target, i.e., b-D-galactose receptors, which are expressed on ovarian cancer, Fig. 35.
The in vivo multicolor imaging was performed in mice bearing ovarian cancer metastases. The excitation of bacteriochlorin with green light results in various NIR fluorescence emissions that may be detected apparently from each fluorophore. The selectivity towards targeted receptors was achieved and also enhanced due to the unique fluorescence properties. The fluorescence emission of bacteriochlorin conjugated to hGSA are normally quenched [285]. Nevertheless, after binding to H-type lectin on the surface of cancer cells, the fluorescence signal was observed showing the tumor localization. The described activity of these compounds was reported only in the case of peritoneal metastases. Based on single excitation, two-color NIR fluorescence imaging with bacteriochlorin-based photoactivatable complexes was developed and reveal that they are worth further investigation [285].
In other studies, Liu et al. have described the bacteriochlorophyll derivative that might be used in multimodal theranostics [288]. The molecular design of this probe consists of bacteriochlorophyll molecule conjugated with a targeting peptide that enables modifying the pharmacokinetic profile as well as recognition of folate receptors on the cancer cells surface. Moreover, this probe indicated significant, red-shifted fluorescence emission  (k ex /k em 748/766 nm). To confirm the selectivity of the prepared probe and the possibility of targeting the folate receptorexpressing cancer cells, its accumulation of tumor tissue was investigated in vivo in KB-and HT1080-tumor-bearing nude mice. In these animal models, KB tumors were characterized by high folate receptor expression and HT1080 with low expression, respectively. Besides the fluorescence imaging, performed studies revealed high in vitro phototoxicity as well as significant photoactivity in vivo resulted in complete tumors regression. Thus, these findings provide that the modified BChl derivative can be a promising dual-mode agent for NIR fluorescence imaging and PDT purposes [288].
In the context of photodynamic treatment, the new frontier is related to selective and controlled light delivery with possible visualization of the administered photosensitizer. Even though the insertion of the optic-fiber into the body and selected internal organs is usually supported and controlled by standard techniques including ultrasonography, optical, or X-ray imaging, the complete and real-time monitoring of the delivered light and the direct impact on tissue response during PDT is currently unavailable [289]. This limitation may be dangerous in the case of tumors localized closely to vital organs, major blood vessels or nerves, because then the light must be spared to avoid the damages or medical complications [289]. In such cases, the precise light delivery during PDT is desirable. For instance, Scherz with collaborators have evaluated the opportunity of the use of blood oxygenation level-dependent (BOLD) contrast MRI for real-time and dynamic monitoring of PDT procedure. They developed this technique using Tookad-PDT against prostate cancer and indicated that during photodynamic treatment, the significant attenuation (ca. 25-40%) of the magnetic resonance signal might be observed only in tumor area subjected to the irradiation [257]. All of the described results reveal that photosensitized BOLD-contrast MRI may be applied intraoperatively as interactive guidance for monitoring antivascular cancer therapy, cardiology, PDT for age-related macular degeneration (AMD) as well as other biomedical purposes [289].
Lanthanide complexes are characterized by strong luminescence in the NIR, with an extremely narrow band, making them ideally suited for use in developing new multifunctional luminescent materials. Unfortunately, these complexes exhibit low molar absorption coefficients due to forbidden f-f transitions. This problem may be omitted by using tetrapyrrolic ligands, characterized by strong absorption and efficient energy transfer from its triplet state, which lay higher than excited state of the lanthanides. The complexation of lanthanide metal ion by tetrapyrrolic ligand enhances the NIR emission and extends the phosphorescence lifetime by narrowing the energy gap between the triplet state of ligand and the excited state of the metal ion (triplet-triplet annihilation upconversion mechanism, TTA-UC, Fig. 36A). The properties of free-base porpholactone ligands and lanthanide(III) metal complexes (Ln = Lu 3+ , Gd 3+ , Yb 3+ ) and their biological activity studies including cellular-uptake as well as photocytotoxicity was performed by Zhang and co-workers [169,170,290]. The comparison of free-base tetrapentafluorophenylporphyrin (F 20 TPP), porpholactone (F 20 TPPL), cis/trans porphodilactone and their Gd 3+ complexes show, that especially in case of trans-isomers the phosphorescence is red-shifted with decreased lifetimes and phosphorescence quantum yields [290]. Moreover, the incorporation of the heavy atom (Gd 3+ ) enhanced photocytotoxicity of these compounds upon red light irradiation that makes them most promising for PDT, Fig. 36B. Furthermore, the presence of Gd 3+ in porpholactone structure allows the rational design of dual-mode photosensitizers due to both therapeutic and diagnostic agents within the same molecular scaffold [290]. Similarly to free-base, the cis/trans isomers of porphodilactones have significant differences in quantum yields of the triplet state of Yb 3+ complexes. Due to this fact, Zhang et al. also constructed a sensitive glucose-oxidase dependent lanthanide poprholactone-based luminescent sensor. In this probe, the NIR emission of Yb 3+ is specifically switched-on by glucose oxidase and then switched-off in the presence of glucose (Fig. 36C). This phenomenon indicates that porphodilactone-based ligands enhance the quantum yield of Ln luminescence and may be used in several biomedical applications, Fig. 36 [290]. The upconversion fluorescence imaging is a promising method due to their biological and technical specification, including excellent photostability, narrow emissions bands, absence of photodamage to living organisms, absence or very low auto-fluorescence, high sensitivity, and high penetration into biological tissues. Upconversion is a non-linear optical process, which involves the conversion of two or more low energy photons (from the NIR) and emits radiation with higher energy [291]. Thus, its potential is commonly used not only in diagnostics but also in PDT [292]. The functionalization of these fluorescent probes may be combined with other imaging methods such as magnetic resonance imaging, positron emission tomography and computed tomography. Porphodilactone complexes with Lu 3+ are suitable for upconversion based on triplet-triplet annihilation [290]. Moreover, they display much longer phosphorescence decay lifetimes (ls scale) than the Pt 2+ , Pd 2+ complexes (ms), and thus can be applied in upconversion-based imaging [169].
Other examples of upconversion probes are nanoparticles and nanocrystals doped with metal ions, especially by f-block metals [291,292]. Moreover, the immobilization of nanoparticles with lanthanide porphyrinoids allows to overcome the major limitation of traditional porphyrin-based photosensitizers for practical application of PDT such as weak absorption in the red region and poor biocompatibility [292]. Similarly to bacteriochlorins, the porpholactones and their metal complexes possess higher cellular uptake than porphyrin, which is crucial for their in vivo evaluation [169]. Furthermore, porpholactones are characterized by a high binding affinity to low-density lipoproteins (LDL). The cytotoxicity tests also reveal that b-lactonization of porphyrin leads to a more prominent photodynamic effect against HeLa cells with photodynamically-induced cell death via apoptosis [169].

Application of bacteriochlorins in photodynamic inactivation of microorganisms (PDI)
The bacteria-related diseases are becoming a serious worldwide health problem due to the increased multi-drug resistance [293]. Although many antibiotics are effective against Gram-positive bacteria, the clinical manifestation of multi-resistant bacterial infec-tions, especially caused by Gram-negative bacteria, is still unsatisfactory. Moreover, microorganisms are capable of developing complex mechanisms of defense against the damages and antimicrobial drugs by, for instance, creating their matrix named biofilm as well as several virulence factors. Thus, the development of new antibacterial strategies is urgently needed [294][295][296][297]. Among them, the photodynamic inactivation of microorganisms (PDI) represents one of the most promising method that involves the excitation of a photosensitizer to generate ROS. Due to the lack of resistance to singlet oxygen, this highly reactive molecule may initiate further oxidative reactions in the nearest environment, such as nucleic acids, lipids, proteins, or enzymes [59,298]. Therefore, PDI is based on the concept that a photosensitizer can accumulate in the cytoplasmic membrane, which is the critical target for damages after irradiation [298], Fig. 37 [299][300][301].
Many classes of photosensitizers were developed for efficient PDI of bacteria, viruses and fungi [157,294,295,298,[302][303][304][305][306][307][308]. It appears that the most appropriate PDI photosensitizers are positively-charged, water-soluble and photostable compounds [309]. Moreover, affinity to bacterial cell wall is a crucial factor which influences PDI efficacy. The differences in morphology between the Gram-negative and Gram-positive bacteria are responsible for PS interaction with the bacterial cell walls. The presence of lipopolysaccharide (LPS) in the outer membrane of Gram-negative bacteria provides an additional barrier to many molecules in the external environment. It contributes to the observed drug-resistance of these organisms. In contrast, Grampositive bacteria contain a single peptidoglycan layer with specific proteins on the cell surface. The PDI effectiveness against fungal yeast is limited by low diffusion of the drug into the fungal cytoplasm because of their cell walls, which also contain b-glucan. Thus, PS with positively charged groups or the addition of substances able to increase the permeability of the outer membrane outstanding enhance the efficacy of Gram-positive bacteria inactivation.
As it has been claimed earlier, the most promising photosensitizers for anticancer PDT are chlorins and bacteriochlorins. Porphyrins, however, can still be considered as efficient agents for antimicrobial approach. Anyway, bacteriochlorins still seem to be more promising for the treatment of biofilm, localized infections and deeper wounds, due to the sufficient NIR light penetration through tissues [310]. Besides the appropriate spectroscopic properties, a photoactive agent should have a high affinity to the bacterial cell wall and allows penetration into bacterial cells [238]. The examples of three mono-substituted cationic bacteriochlorins with quaternary ammonium substitution (named BC37, BC38, BC39, Fig. 37. Scheme of photodynamic inactivation of microorganisms (PDI) and its basic mechanims of action. Fig. 38.) were synthesized and tested as antimicrobial photosensitizers [311].
High effectiveness at nanomolar concentrations and low light dose (10 J/cm 2 ) against Gram-positive bacteria was reported for bacteriochlorins presented in Fig. 38, with a different number of positively-charged groups. The bacteriochlorin bearing two cationic moieties indicates the highest activity against the resistant bacteria (at micromolar concentrations), whereas the hydrophobic one was more effective against the fungi [311]. According to these studies, it can be concluded that Gram-positive bacteria are more susceptible to PDI by neutral photosensitizers or those with a moderate number of positive charges with relatively low hydrophobicity [311]. In contrast, effective photoinactivation of Gram-negative bacteria requires more hydrophobic photosensitizers with many positive charges in the structure [303]. In turn, fungal cells should be treated with photosensitizers that have little or no positive charge with relatively higher hydrophobicity [311].
It should also be highlighted that efficient photodynamic inactivation of a broad-spectrum of pathogens can be achieved by a protocol involving short photosensitizer-cell incubation time (up to 2 h) and relatively low light dose (i.e., 10 J/cm 2 ). Thus, in the following studies, the authors also showed that a short incubation time (up to 30 min) might also result in increased selectivity for microbial cells over human cells (HeLa cancer cell line) [311].
The possibility to use shorter incubation times in PDI than for typical photodynamic effect against cancer or eukaryotic cells is a consequence of the fast-electrostatic interaction occurring between cationic functional groups in PS structure and the negatively charged teichuronic and lipoteichoic acids located in the outer wall of bacterial and fungal cells [303].
In other studies, Hamblin with coworkers tested the next set of bacteriochlorins containing a various number of cationic charges against the different species of microorganisms, Fig. 39 [238]. It was reported that all four derivatives indicate the PDI activity against Gram-positive S. aureus. The most significant antibacterial effect was achieved for the bis-quaternized derivative (5 logs reduction after PDI with 100 nM PS). However, other bacteriochlorins (basic, tetrakis-as well as hexakis-quaternized) showed similar but relatively lower than the best one, activity in bacteria inactivation procedure (up to 5 logs of killing after PDI with 1 mM PS) [238]. The observed effect may be explained by the statement, that there is an optimum number of cationic charges in the photosensitizer structure, which is responsible for: (i) PS binding to anionic phosphate groups present on the bacteria cell wall and (ii) efficient penetration into the bacterial cell wall, where ROS are produced upon irradiation leading to photodamages. The lower than the optimal number of positive charges (for instance, neutral bacteriochlorin derivative with two basic amino groups) dimin-  ishes the sufficient binding. In contrast, the respective higher number of cationized groups like in bacteriochlorin containing 4-6 positive charges) result in too strong binding, which reduced the PS penetration into the cell wall [238]. A similar finding has also been reported for PDI with chlorin e6 conjugates with different-sized poly-L-lysine [312,313] or polyethylenimine [313] chains as well as porphyrin derivative [157,309,314]. Maisch et al. have confirmed that porphyrin-based photosensitizer with two cationic groups indicated higher activity against S. aureus than an analog molecule with four positively-charged substituents [314].
The observed results for cationic bacteriochlorins were also supported by the photophysical studies and DFT calculation undertaken in the following studies [238]. The theoretical analysis revealed that the excited triplet state quantum yield, which is the precursor of ROS, for the four bacteriochlorins is similar and reached ca. 0.48-0.53 [238]. Moreover, any correlation between the redox properties of photosensitizers was observed [238]. The only noticeable trend is that mono-substituted bacteriochlorins are more effective than di-substituted derivatives, which may suggest that their photoactivity is associated with an increased probability of the electron transfer processes [238]. Thus, it can be suggested that overall PDI activity observed among the set of bacteriochlorins must be related to both cellular binding and accumulation rather than photochemical features. All of these data present the basic information about the structure-activity relationship for analyzed cationic bacteriochlorins and indicate the major factors that influence their photodynamic activity [238].
Other authors also investigated the bacteriochlorin-based PDIphotosensitizers. For instance, Meerovich et al. have also studied the effect of cationic charges in the photosensitizer structure. It appeared that the investigated tetracationic photosensitizers, based on synthetic bacteriochlorins with reduced molecular size and molecular weight, were highly efficient in photodynamic inactivation of Gram-positive bacteria (S. aureus) and Gram-negative bacteria (P. aeruginosa) and against their biofilms [315]. The E. coli photoinactivation profile by cationic isobacteriochlorin described by Faustino et al. also proved that by using adequate PS, the PDI could be used in clinical applications. The results obtained for the series of modified isobacteriochlorins confirm that the trend of inactivation efficiency proceeds with the increasing number of cationic groups in the PS structure. The synergistic effect of red-shifted absorption and a positive charges in investigated isobacteriochlorin led to a 6 logs reduction and indicated that cationic isobacteriochlorins seem to be a promising agents for photodynamic inactivation of Gram-negative bacteria.
The future outlook of the treatment of resistant pathogens is the design and development of hybrid materials with multiple mechanisms of action. For these reasons, nanotechnology-based strategies in combination with PDI are more frequently explored [295,304,305]. Properly designed hybrid nanosystems may increase the affinity to microorganisms, enhanced ROS generation as well as are able to overcome the multidrug resistance mechanisms. The last one seems to be most important, due to the fact that the present time is named as the ''end of the antibiotic era" [294]. This motivated a search for alternative antimicrobial strategies, including PDI to overcome multidrug-resistance mechanisms [294,302].
8. Other applications of bacteriochlorins beyond the medicine 8.1. Application of (metallo)bacteriochlorins conjugates as a lightharvesting antennas The primary function of metallobacteriochlorins is to participate in the photosynthesis reactions consisting of, among others, the NIR photons absorption that leads to charge separation and efficient transformation of solar energy into chemical energy in the forms of sugars that later serve as fuel for leaving organisms. Therefore, an increasing number of studies are being conducted on the use of similar compounds as the potential light-harvesting antenna (LHA) models. LHA that captures light energy and then funnels excitation energy to the reaction center is very important for natural photosynthesis. Among the reported natural or artificial LHAs, porphyrins or their derivatives have been widely used as donors or acceptors. However, the absorption spectra of porphyrins cover only a limited portion of the visible region, which is a big barrier to light harvesting. The design of the bioinspired light-harvesting devices using well organized supramolecular arrays must be considered as one of the most promising system. The requirements for ideal supramolecular assemblies are i.e., strong absorption, efficient solar energy storage and easily transferred without energy loss [316]. Metal-coordinated chromophores assemblies are widely studied for antenna systems due to their large molar absorption coefficients. In addition, metal ion coordination ensures a strong impact through ion interaction as well as through the formation of a hydrogen bond [316]. The construction is based on metallotetrapyrolles linked with free-base molecule by linker with different hindrances and in different positions, which provide changes in the HOMO-LUMO gap. Excellent energy transfer and weak electronic coupling may be obtained by using such a linker because the distance between linked chromophores as well as a hindrance is crucial to ensure efficient excitation and energy transfer.
In the systems containing Mg 2+ or Zn 2+ complexes as energy donors and free-base acceptor, the energy transfer process is characterized by short lifetime (s~25-115 ps) and high quantum yield (U~95%) [317]. Another class of extensively studied artificial antennas is encompassed synthetic tetrapyrroles, likely zinc bacteriochlorin-bacteriochlorin (ZnBC-BC) complexes. Shoji, Nomura, and Tamiaki have investigated covalently linked heterodimers of porphyrin, chlorin, or bacteriochlorin with or without zinc ion [124]. They have also compared the axial ligation of Zn 2+ in porphyrins, chlorins and bacteriochlorins, concluding that the more flexible and less p-conjugated bacteriochlorins bind axial pyridine stronger than the corresponding chlorins or porphyrins. They have reported several Zn 2+ bacteriochlorins that, upon selfaggregation, exhibit significant red-shifts in their absorption spectra [318][319][320]. Based on naturally-occurring photosynthetic systems with a large number of excitonically coupled tetrapyrrolic arrays or strongly conjugated by a linker (double and triple C-C bonds) with faced model assembly were investigated [321] Fig. 40 [117,122,322].
Ptaszek with co-authors have found out that the dimer of chlorin (cis-2C), bacteriochlorin-chlorin (cis-2BC), and zincbacteriochlorin-chlorin (cis-2ZnC) linked via enediyne bond exhibited thermally stable isomers. The slipped co-facial arrangement of the macrocycles, in which the structure may be stabilized through space-and bond-interactions were studied [40]. Nevertheless, the computational of the trans arrangement along linkage double bond has no splitting. NMR spectra and DFT calculations predict the antiparallel arrangement of macrocycles. In addition, theoretical calculation confirms that the anti-parallel configuration of chromophores provides the lowest energy for cis configuration of the linking bridge. The introduction of Zn 2+ into the system indicates strong excitonic coupling in contrast to the identical free-base supramolecular arrangement. The excited state of cis-2ZnC tends to decay with reduced fluorescence. Additionally, the dimers of the tetrapyrrolic core possess lower fluorescence quantum yield (U F ) compared to their monomers. This study shows that the geometry and spectral properties of the investigated dyad are similar to those observed in particular pairs involved in photosynthetic reaction centers [40]. Thus, they may be constructed components of energy and electron transfer for modeling the photosynthetic reaction pathway. In addition, the chromophores coupled with enediine connector, able to subsequent photoisomerization to trans-cis conjugates, seems to be useful elements for designing the energy conversion systems and photonic materials with new optical properties.
In the co-assembly model containing an aggregative zinc 3 1hydroxy-13 1 -oxo-chlorin and free-base bacteriochlorin with appropriate energy level, the covalently linked BChl heterodimer was arranged in chlorosomal chlorin J-aggregates. In particular, co-assemblies of porphyrin J-aggregates with heterodimers were first developed, and the J-aggregates with zinc-chlorin with bacteriochlorin (ZnC-B) or zinc-porphyrin with bacteriochlorin (ZnP-B) exhibited relatively higher intense emission than the free-base bacteriochlorin conjugated with chlorin. The Zn 2+ complexes possess a particularly narrow HOMO-LUMO gap. Thus, the intramolecular photoinduced electron transfer processes to the nearest molecule can easily occur. Zinc-bacteriochlorin linked by the alkyne linker group with the sterically unhindered alkyne linker (strong interactions) attached in the different sites is characterized by a strongly reduced in the HOMO-LUMO gaps, and low tetrapyrrolic-linker rotational barriers, Fig. 41 [123,323].
In contrast, bulky aryl linkers have limited interaction between linked macrocycle resulting in increased HOMO-LUMO gap energy and higher ZnBC-linker rotational barriers. The design of supramolecular conjugates based on linker interaction is important for development of efficient light-harvesting arrays [324]. Computational studies of these systems are very feasible and can be helpful for further prediction of their application as a components of the antenna system. Okuno and Mashiko have proposed the mechanism of the interaction with different linkers and chromophores in porphyrin arrays. The comparison of excitation energy with rotational energy barrier for Zn-porphyrin shows strong dependence, in contrast to the free-base analog, in which the HOMO-LUMO energy gap is independent of p conjugation [324].

Application of bacteriochlorins as chemical sensors
It is commonly recognized that optical sensors relying on fluorescence detection indicate higher sensitivity than absorptionbased ones. For fluorescent optical sensors with optimized performance and sensitivity, the use of a single-excitation and dualemission fluorophore allows for ratiometric signal processing. It is essential to overcome some disadvantages (photobleaching, indicator leaching, light source intensity fluctuations, etc.) of single wavelength fluorescence intensity approaches. Suzuki, Tamiaki et al. have described the use of bacteriochlorin derivative as a fluorescent ratiometric optical, chemical sensor (optode) for alcohol detection [325]. In previous studies, the same authors reported the synthesis of a trifluoroacetylchlorin as the first example of a fluorescent probe for alcohols and amines using a chlorophyll derivative as the central dye unit.
Due to the fact that optical sensors with spectral sensitivity in the long-wavelength range are less prone to optical interference (light absorption, autofluorescence), the authors appllied a lipophilic derivative of 8-oxo-bacteriochlorin incorporated into a plasticized poly(vinyl chloride) membrane. With this fluorophore, sensitive and reversible ratiometric ethanol sensing in the NIR range with reversible signal changes was achieved. Immobilization of the fluorophore into a plastic membrane of the poly(vinyl chloride) dramatically change both, their absorption and emission properties. The blue-shifted of fluorescence from 750 nm to 701 nm in the presence of ethanol may be used as an ethanol sensing probe for beverage and industrial analyses [325].

Bacteriochlorin based organic-solar cells for efficient energy conversion
Metal complexes with tetrapyrrolic ligands, due to appropriate excitation energy and high molar absorption coefficients in Vis/ NIR, are particularly interesting in terms of their possible application in dye-sensitized solar cells (DSSC). The functioning of DSSCs depends on the efficiency of the electron injection from a photoexcited sensitizer molecule to the conduction band of the semiconductor, yielding mobile electrons and the oxidized dye. The electron injection and recombination of separated charges at the sensitizer-semiconductor interface are the crucial processes influencing the efficiency of solar energy conversion [326,327]. Metallotetrapyrroles, mainly with coordinated Zn 2+ , were successfully used to sensitize TiO 2 in DSSC [328]. Due to the strong absorption within the green (ca. 500-550 nm) and NIR regions of spectra, the bacteriochlorin derivatives are indicated as versatile molecules for the construction of photovoltaic devices. So far, the most popular DSSCs are these containing ruthenium complexes adsorbed at the surface of nanocrystalline TiO 2 to collect photons up to 650 nm. However, their main limitation is the energy conversion efficiency of 11%. Extending the absorption to 800 nm could increase the maximum global solar radiation from 24% to 40%.
Organic solar cells are a prospective alternative for classical solar energy conversion into electrical energy due to the effective heterojunction of electrons in the active layer [329,330]. Among the several types of polymers or small molecules based on donor-acceptor, the conjugates with bacteriochlorin as a core with two electron-withdrawing terminal units linked by two p-linkers, are characterized by remarkable photovoltaic productivity. However, power conversion efficiency (PCE) of the OSCs depends on the strong absorption of the solar spectrum region and efficient electron transfer by the active layer. In this context, Ponsot and colleagues have designed a molecule based on the electro-donor central core of bacteriochlorin linked with two different electronacceptors groups via p linker bridges for heterofunctional organic solar cells [330]. The use of small-molecule ligand as a complementary donor enables to increase of the efficacy of this binary system up to 9.88%, which is probably related to a deeper HOMO ligand energy level. In addition, the respective LUMO offset reaches only~0.30 eV, which is very similar to the threshold value necessary for exciton dissociation and charge transfer in mass group substitution causes heterojunction to the active layer with minimal energy loss [330]. The authors indicated that the bacteriochlorin with a more extended p-system exhibits a narrow HOMO-LUMO energy gap, which is desirable properties for lightharvesting in the NIR region of the solar spectrum [330].

Summary and conclusions
Near-infrared (NIR) part of the electromagnetic radiation, despite its advantages, has been relatively poorly used in medicine, optoelectronics and photocatalysis due to the lack of thermally and photochemically stable chromophores, that strongly absorb electromagnetic radiation of this range. Bacteriochlorins are an exceptional example of NIR-absorbing compounds. They have been chosen by nature to conduct photosynthesis without the produc- tion of oxygen in various phototropic bacteria. The reason for this selection is that these molecules efficiently absorb photons in the range of 700-900 nm and have appropriate redox properties. The best-known example of bacteriochlorin is bacteriochlorophyll a, a natural pigment structurally similar to chlorophyll a, but containing two reduced pyrrole rings. This is the main reason why its naturally occurring derivatives are characterized by reduced stability and high cost of production. Therefore, in recent years there has been an increased interest in obtaining rationally designed synthetic bacteriochlorins, with more favorable photochemical properties and increased stability. The absorption spectra of bacteriochlorins reveal an extraordinary high absorption in the infrared (e~100,000 M À1 cm À1 ), a very high absorption in the near UV, and relatively low absorption in the green part of the visible light. The stability of synthetic bacteriochlorins is assigned to the presence of desired metals ions, electron-withdrawing groups (e.g., cyano, geminal dimethyl and halogen substituents) and steric protection. The presence of such substituents in the bacteriochlorin structure results in a marked increase in oxidation potentials and thus, in a lower tendency of these compounds to photodegradation. The proper control of oxidation potentials is also important for the determination of ROS generation mechanisms, and more precisely, for the engineering of photoinduced electron transfer and energy transfer processes.
The possibility of introducing various functional groups and metal ions into the bacteriochlorin core for fine-tuning of optical properties and polarity and the increased possibility to obtain large quantities of substantially pure metallobacteriochorins led to a wide range of practical applications. They are useful antennae systems for light-harvesting in the near-infrared (NIR) and are considered as effective agents in the photodynamic therapy (PDT) of cancer, photodynamic inactivation of microorganisms (PDI), in vivo oxygen imaging, fluorescent labeling of biomolecules, photodiagnosis (PD), solar energy conversion, optical sensing, and photocatalysis. The recent results from various scientific groups, including our, have shown that bacteriochlorins are promising compounds for PDT because they can use NIR radiation that deeply penetrates through the tissues and convert it very effectively to highly reactive oxygen species such as singlet oxygen, superoxide ion, hydrogen peroxide and hydroxyl radicals. In vitro studies with several families of (metallo)bacteriochlorins generally revealed their low cytotoxicity in the dark and high phototoxicity to various cancer cells. In a few independent studies, it was demonstrated that the photodynamic effect does not depend on the quantum yields of singlet oxygen formation, but it is strictly correlated with the production of oxygen-centered radicals. Bacteriochlorins that generate free radicals more efficiently lead to a more effective photodynamic effect than photosensitizers generating only singlet oxygen. The experiments performed on different animal models showed favorable pharmacokinetics, biodistribution and photodynamic efficacy. It was also demonstrated that the functionalization of bacteriochlorins may increase the effectiveness of the applied therapeutic protocols with polymeric micelles, lipoproteins, nanoparticles and metal-organic frameworks.
The best example of a bacteriochlorin containing metal ion, that has been successful in clinical trials, is Tookad Soluble (padeliporfin). After passing stage III and undergoing phase IV clinical trials, it is finally approved in the European Union and in Mexico for the treatment of low-risk early prostate cancer. Another bacteriochlorin-based photosensitizer discussed in this work, which is currently in Phase I/II of clinical trials, is redaporfin. Redaporfin is a photosensitizer from the family of halogenated bacteriochlorins that absorbs NIR photons efficiently and convert them into cytotoxic ROS. Its efficacy in the treatment of various types of tumors, including lung, colon, prostate, breast and melanoma has already been confirmed in preclinical tests. The promising results of Phase I/II clinical trials for the treatment of advanced head and neck cancer clearly reinforce the beneficial results of non-clinical studies. PDT with redaporfin as photosensitizer offers a number of advantages over existing anticancer regimes, such as improved efficacy, good tolerability, ease-of-use, and low-cost efficiency. Some of (metallo)bacteriochlorin described herein are also characterized by an intense fluoresce at 750-900 nm, where tissues are most transparent, and a large Stokes shift, offering opportunities for photodiagnosis and for their use as sensors. Needless to say, the development of NIR-absorbing materials for therapy and diagnosis may have a large impact on health care. Ongoing research in this area may lead not only to the use of novel compounds with increased performance in currently available treatments but also to the discovery of innovative therapeutic approaches.
Novel stable bacteriochlorins and their metal complexes are constantly and comprehensively characterized by advanced spectroscopic, photophysical, photochemical, electrochemical, and biochemical techniques. Hybrid materials obtained as a result of their combination with semiconductors and other nanostructures offer new interesting photochemical, photophysical, and redox properties. Smart control of these features is possible by the choice of proper metal, ligand modification, supramolecular architecture and anchored nanoparticles. Clearly, stable (metallo)bacteriochlorins have opened a new window to the NIR region of the electromagnetic spectrum, which has never before been so widely exploited in chemistry, biology, medicine and industry.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.