Germanium

Endpoint summary

Description of key information

Acute Dataset:

Four acute freshwater studies were performed using GeO2 as test substance and Danio rerio (fish), Daphnia magna (invertebrate), Rhaphidocelis subcapitata (micro green algae) and Navicula pelliculosa (diatom) as test species; or Ge Atomic Absorption standards as the test substance and the sediment dwelling freshwater aquatic species Hyalella azteca (Table A). The most sensitive species was the diatom N. pelliculosa with an EC50 of 0.206 mg GeO2/L, or 143 µg Ge ion/L.

 

Table A. Aquatic acute dataset for Germanium.

Trophic Level	Species	EC/LC50 (mg GeO2/L)	EC/LC50 (mg Ge/L)	End point	Ref.
Algae	N. pelliculosa	0.206	0.143	72-h Growth Rate	Fraunhofer 2018
	R. subcapitata	178	123.6	72-h Growth Rate	LISEC 2001
Invert. inc. sediment	D. magna	67.5	46.9	48-h Immobilisation	LISEC 2001
	H. azteca*		>3.1	7-d mortality, tap water	Borgmann et al. 2005
	H. azteca*		0.209	7-d mortality, soft water	Borgmann et al. 2005
Fish	D. rerio	103.5	71.8	96-h mortality	LISEC 2001

* sediment species, but with exposure in the water column; “Atomic Absorption Ge standards” used as test substance, not GeO2. A low LC50 was observed only in very low hardness (18mg/l) water.

 

In addition, several bacteria and yeasts were also tested (Van Dyke et al., 1989); effects were seen at high concentration (100 mg GeO2/L and more), only.

 

Chronic Dataset:

There were only two freshwater studies out of a total of 24 chronic studies. One freshwater study species was one of the most sensitive to Ge, whereas the other species was the most tolerant. As freshwater species sensitivity was comparable to marine species (particularly with regards to the sensitive species) the datasets were combined into a single chronic dataset. In combination, there is a total of 24 chronic EC10/NOECs from 19 different species of algae included in the chronic dataset (Table B;17 diatom studies on 12 different species, 1 study on a single micro-algal species; and 6 studies on 6 different macro-algae species).

The data available in the open literature focus on the taxonomic group of the diatomeae which is exceptionally sensitive to Ge. Indeed, it is since long known that in diatoms, with their Si-based exoskeleton, Germanium can substitute for Si and as such cause toxicity (Mehard et al., 1973; Azam, 1974). This unique toxicity mode of action amongst taxa explains why diatoms are so sensitive to Ge. For further discussion, see section “Why are diatoms the most sensitive taxonomic group?” below.

The open literature provided some extensive studies on marine diatomeae. These studies, albeit of good scientific quality, were not performed under standard conditions or similar. Therefore, an additional study was performed on a freshwater diatomea, according to OECD 201 standard conditions.The most sensitive species in the combined dataset was the marine diatom species T. rotula; however, this value is of low quality, since this NOEC was estimated graphically from a dose-response curve, and no statistics are given. The value is included in the SSD, but is not considered reliable for deriving the ecotoxicty reference value (erv). For the chronic erv, the second most sensitive result is used, obtained on the freshwater diatom N. pelliculosa , with an EC10 of 100.3 µg GeO2/L, or 70 µg Ge ion/L, obtained by an OECD 201 test.

 

Table B. Chronic ecotoxicity data for Germanium

 

Species		EC10/ NOEC	EC>10?	ECx [Ge](mg GeO2/L)	ECx >10 to NOEC	NOEC/EC10 (mg GeO2/L)	GeO2 to Ge (mg Ge/L)	Endpoint	Ref.
T. rotula	Diatom SW	EC10*			na	0.05	0.031	4d Cell no.	Markham & Hagmeier 1982
N. pelliculosa	Diatom FW	EC10			na	0.10	0.070***	72h G. rate	Fraunhofer 2018
N. longissima	Diatom SW	NOEC			na	0.13	0.093	4d Cell no.	Markham & Hagmeier 1982
Amphiprora	Diatom BW		EC50	1	÷ 5	0.20**	0.139	12d Growth	Lewin 1966
C. fusiformis	Diatom SW		EC50	1	÷ 5	0.20**	0.139	12d Growth	Lewin 1966
C. fusiformis	Diatom SW		EC50	1	÷ 5	0.20**	0.139	12d Growth	Lewin 1966
P. tricornutum	Diatom SW		EC50	1	÷ 5	0.20**	0.139	12d Growth	Lewin 1966
S. costatum	Diatom SW		EC50	1	÷ 5	0.20**	0.139	12d Growth	Lewin 1966
F. spiralis	Macro-algae SW	EC10*			na	0.22	0.153	6d length.	Markham & Hagmeier 1982
L. saccharina	Macro-algae SW	NOEC*			na	0.22	0.153	6d length.	Markham & Hagmeier 1982
C. fusiformis	Diatom SW		EC30	1	÷ 3	0.33**	0.231	12d Growth	Lewin 1966
Navicula sp.	Diatom SW		EC20	1	÷ 2	0.50**	0.347	12d Growth	Lewin 1966
N. angularis	Diatom SW		EC20	1	÷ 2	0.50**	0.347	12d Growth	Lewin 1966
N. incerta	Diatom BW		EC20	1	÷ 2	0.50**	0.347	12d Growth	Lewin 1966
C. closterium	Diatom SW		EC50	3	÷ 5	0.60**	0.417	12d Growth	Lewin 1966
C. fusiformis	Diatom SW		EC30	3	÷ 3	1.00**	0.694	12d Growth	Lewin 1966
P. tricornutum	Diatom SW	EC10		1	÷ 1	1.00	0.694	12d Growth	Lewin 1966
U. lactuca	Macro-algae SW	NOEC*			na	1.10	0.764	6d Cell diam.	Markham & Hagmeier 1982
C. concinnus	Diatom SW	NOEC			na	1.79	1.24	6d Cell no.	Markham & Hagmeier 1982
N. pelliculosa	Diatom FW	NOEC*			na	2.00	1.388	7d Growth	Lewin 1966
C. crispus	Macro-algae SW	NOEC			na	8.95	6.21	6d Growth	Markham & Hagmeier 1982
P. urcelotata	Macro-algae SW	NOEC			na	8.95	6.21	6d Growth	Markham & Hagmeier 1982
P. umbilicus	Macro-algae SW	NOEC			na	8.95	6.21	6d Growth	Markham & Hagmeier 1982
R. subcapitata	Micro-algae FW	EC10			na	73.00	50.68	72h G. rate	LISEC 2001a
Daphnia magna	Daphnia  FW	EC10				0.544	0.378	offspring per survived parent	Fraunhofer 2020

 

 

legend:

·       ‘SW’: seawater, ‘FW’: freshwater, ‘BW’: brackish water.

·       *: Not listed in the paper as an EC10 or NOEC, but estimated graphically from a dose response curve; no statistics. 

·       ‘EC>10?’: For Lewin 1966, the % of control affected by the effect concentration, rounded up to an EC. See text for details.

·       ‘ECx [Ge]’: The concentration at which the effect to algal growth was observed in the Lewin 1966 study (essentially a LOEC). See text for details.

·       ‘ECx > 10’ were converted to a NOEC using EFSA formula.

·       ** NOEC recalculated from ECx. 

·       ***: Most reliable, GLP lab, lowest chronic value.

 

Given the extent of the chronic data, the PNEC was derived by statistical analysis, using all the chronic data in a species sensitivity distribution (SSD). To populate the SSD, the preferred toxicity endpoints are EC10’s. However, as many chronic studies only present the NOEC, this was also accepted. In some cases, NOECs or ECx were not reported, but they could be estimated from the dose-response curves graphically. These values were included in the SSD, but not considered sufficiently reliable for derivation of the ecotoxicity reference value (Erv) for classification.

In the older study by Lewin (1966), data were not presented as ECx’s or NOECs, but the quantitative information on the toxicity data still made it possible to derive an EC10 or NOEC (see below).

 

Converting an ECx>10 to a NOEC:

The study by Lewin (1966) is quite important, because it presents Ge effects on a wide range of diatomeae, under environmentally relevant conditions. However, results were presented in a non-standard way: he presented the percentage of growth rate in a GeO2 exposed population when compared to the control, with the control being 100 %. For example, when N. angularis was exposed to 1 mg GeO2/L (column entitled‘Effect [Ge]’;Table B), the growth rate of the diatom was 80 % of the control. This percentage was then converted to an ‘Effect Concentration’, i.e. the concentration at whichX% was affected. So, using the same example, exposure to 1 mg GeO2/L resulted in a 20 % growth reduction in N. angularis when compared to the control; therefore 1 mg GeO2/L was effectively an EC20 for N. angularis.

The next step involved converting an ECx (with 10<x<50) to a NOEC. There are no specific guidelines for this, but reference could be made here to the EU risk assessment made on zinc under the EU “existing substances regulation”, where NOECs were derived from EC20 and EC30 values (JRC 2010). In this EU risk assessment, it was considered that EC20s were a factor of 2 higher than the NOEC, and EC30s were a factor of 3 higher than the NOEC. A study by EFSA examined the relationship between EC’s and NOECs for 70 plant protection substances, across 615 separate studies on many different organism (including algae), and found identical results. However, they took the relationship up to EC50s, where it was shown that EC50s were a factor of 5 higher than the NOEC (Azimonti et al., 2015). These corrections were applied to the ‘Effect [Ge]’ in the chronic dataset. Therefore, to complete the example for N. angularis, the EC20 was 1 mg GeO2/L, so to convert to a NOEC this concentration was divided by 2: 1/2 = 0.5 mg GeO2/L NOEC for N. angularis.

  

Why are diatoms the most sensitive taxonomic group?

It has long been known that diatoms substitute the uptake of silicon with germanium (Mehard et al., 1973; Azam, 1974). The requirement for Si is unique to the diatomeae. Ge acts by inhibiting silicon metabolism. Specifically, Ge blocks Si uptake, which results in blocking of cell division by blocking cell wall formation (Markham and Hagmeier 1982). This unique toxicity mode of action amongst taxa explains why diatoms are so sensitive to Ge. This difference in sensitivity is also documented in the acute database, where the EC50 for the freshwater diatom N. pelliculosa is much lower than the other EC50 values in the dataset.

The presumed mechanism of Ge-action is via direct competition at the site of uptake. The Ge:Si ratio in natural freshwater and seawater environments is approximately 1:400,000 (FOREGS, 2005 and Sutton et al. 2010, respectively). Therefore, in the natural environment Si easily outcompetes Ge for diatom uptake. However, in the aforementioned ecotoxicity tests (Table A and B), the Ge concentrations applied in the tests are orders of magnitude higher than what is found in the environment (‘mg/L’ in toxicity tests vs. ‘ng/L’ in the environment); yet the silicon concentrations are similar. This creates Ge:Si ratios in the toxicity test of approx. 1:0.5 up to a maximum of 1:8 (Lewin 1966, and Markham et al. 1982). This is obviously a massive gradient shift, ultimately resulting in elevated concentrations of Ge being taken up by the diatom. For these reasons, the effects assessment of Ge substances was focused on the exceptionally sensitive taxonomic group of the Diatomeae. This group is well documented in the present report.

Additional information

Marine and freshwater ecotoxicity data were merged because no difference in sensitivity was observed between them.